Cognitive Apprenticeship and Situative Theory in Engineering Capstone Design: A Systematic Review of Pedagogical Mediators of Professional Identity at the Threshold of Practice

Systematic Review Protocol — Draft for Co-Development with Jentry Campbell, Dartmouth Engineering Library

Authors
Affiliations

Emily Monroe

Thayer School of Engineering, Dartmouth College

Jentry Campbell (proposed)

Dartmouth Library

Eden Gray (proposed)

Dartmouth College

Published

April 2026

1 Purpose and Positioning

This protocol specifies a systematic review of the peer-reviewed empirical literature on how cognitive apprenticeship (Collins, Brown, & Newman, 1989; Collins, Brown, & Holum, 1991; Collins & Kapur, 2014) and situative / situated learning theory (Lave & Wenger, 1991; Wenger, 1998; Brown, Collins, & Duguid, 1989; Greeno, 1998) have been operationalized within undergraduate engineering capstone design courses to shape students’ professional identity as engineers.

The review sits inside the dissertation Designing Safer Systems for Humans: From Engineering Education to Public Health Interventions. That dissertation begins from a premise drawn directly from the licensed engineer’s oath — to hold paramount the public’s safety, health, and welfare — and asks how undergraduate engineering education can form practitioners who can actually discharge that obligation in decentralized, rapidly-changing technology contexts.

Capstone design is positioned at the liminal boundary between undergraduate preparation and early-career practice — the space it occupies structurally in the ABET-accredited curriculum. That boundary is now being reshaped by three classes of hazard, all of which engineering graduates will encounter on their first day of work:

  1. Environmental and public health exposures embedded in engineered systems — ultrafine particles and VOCs from desktop 3D printers in K–12 and makerspace settings, indoor air quality in schools and workplaces, ventilation failures during respiratory pandemics, and the diffusion of fabrication technology into populations without professional oversight.
  2. AI-driven engineering work, where generative and predictive models increasingly participate in design, analysis, verification, and decision-making — creating new demands for human-in-the-loop judgment that rests on professional identity, not just technical skill.
  3. Cyber-physical exposure across every sub-discipline — from operational technology in energy infrastructure to firmware in consumer and medical products to building automation and HVAC controls, and including the use of 3D printers to produce untraceable firearm components or medical countermeasure devices.

These are three faces of the same underlying question: how does engineering education form the professional judgment that makes engineers reliable safeguards of public welfare? Public and environmental health is the dissertation’s empirical anchor — the SHIELD-3D cluster-randomized trial measuring FeNO and 1-OHP biomarkers in school children, the AirGradient field study of VOC/PM2.5/UFP exposure, and the Taguchi L18 chamber study of 3D printer emissions together test whether engineered controls actually protect people. AI verification and cyber-resiliency are the forward-looking applications of the same logic: the same professional identity that lets an engineer say “this ventilation system is not adequate for the exposure” is what lets that engineer say “this AI-generated design has not been verified” or “this firmware update has not been security-reviewed.”

The review will therefore serve three linked audiences: (a) engineering education researchers seeking theoretically grounded evidence on capstone as a site of identity formation; (b) capstone instructors and program directors designing interventions under resource and accreditation constraints; and (c) the Professional Formation of Engineers (PFE) community connecting capstone pedagogy to PE licensure and to emerging questions about who takes responsibility for the public-facing consequences of engineered systems.

The work is conceived as a standalone contribution to a journal like Engineering Education Review (review article, >5,000 words, >20 references, APA) and as the theoretical grounding chapter for the dissertation’s capstone intervention study.

2 Research Questions

2.1 Primary Question

RQ1. How have cognitive apprenticeship and situative theory been operationalized, together or separately, within undergraduate engineering capstone design courses, and what does the empirical evidence indicate about their effects on students’ professional identity as engineers?

2.2 Secondary Questions

  • RQ2. Which pedagogical mediators from the two frameworks are reported as most active in capstone, and under what structural conditions (team size, sponsor type, advisor model, single-discipline vs. interdisciplinary)?
  • RQ3. How is professional identity operationalized and measured in the capstone literature, and how adequate are the measures for capturing identity at the transition to practice — including the emerging responsibilities of verification, public-health-aware design, and cyber-physical stewardship?
  • RQ4. How do the studies treat the transition to engineering work, including licensure pathways (FE/PE) and entry into sectors where public-facing consequences (health, safety, infrastructure, AI) are central?

2.3 PEO + SPIDER Framing

Because the evidence base will include quantitative, qualitative, and mixed-methods studies, a hybrid PEO + SPIDER frame is more appropriate than strict PICOS:

Element Specification
Population Undergraduate engineering students (any discipline) enrolled in a capstone design course or culminating design experience
Exposure / Phenomenon of Interest Pedagogical design, instructional moves, or learning environments explicitly grounded in cognitive apprenticeship and/or situative / situated learning / communities of practice
Comparator (where applicable) Traditional instruction, prior cohorts, cross-sectional comparisons, or pre/post designs
Outcome / Evaluation Professional identity, engineering identity, professional formation, identity trajectory, professional socialization, readiness for practice, or proxies (self-efficacy, role identification, affiliation)
Study Design Empirical studies: quantitative (experimental, quasi-experimental, correlational), qualitative (phenomenological, narrative, ethnographic, case study), and mixed-methods
Research Type Peer-reviewed journal articles and peer-reviewed conference proceedings (ASEE, FIE, REES, CDIO, CEEA, SEFI)

3 Why This Review Is Needed: Gap in the Literature

A preliminary scoping of adjacent reviews confirms a defensible, previously unfilled space:

  • Pierrakos, Pappas, & Holcomb (2024, IJEE) offered a reflective account of cognitive apprenticeship applied to Wake Forest Engineering’s capstone. It is a single-institution conceptual piece, not a systematic review, and does not incorporate situative theory or professional identity measurement.
  • Minshew, Olsen, & McLaughlin (2021, AERA Open) systematically reviewed cognitive apprenticeship in STEM graduate education. Their sample excluded undergraduate capstone and was not focused on identity as the primary outcome.
  • Morelock (2017, European Journal of Engineering Education) systematically reviewed engineering identity but was not structured around specific pedagogical theories and treated capstone as one context among many rather than as the primary unit of analysis.
  • Tenhunen, Männistö, Luukkainen, & Ihantola (2023, Information and Software Technology) systematically reviewed software engineering capstones but focused on course characteristics with no theoretical lens on identity formation.
  • Heitzner, McKay, & Hurst (2025, CEEA Proceedings) reviewed capstone assessment schemes — a methodological adjacency, not a theoretical or identity-focused treatment.
  • Donaldson, Fore, Filippelli, & Hess (2020, International Journal of Science Education) systematically reviewed situated learning — but in geoscience, not engineering, and not at the capstone stage.
  • Parker (2019, doctoral dissertation, UC Boulder) applied situated learning theory to a CS capstone for professional identity but is a single-site study, not a synthesis.
  • Johri & Olds (2011, Journal of Engineering Education) provided the foundational theoretical argument for situative learning in engineering education; it was a scoping/theoretical piece, not a systematic evidence synthesis.

The gap: No systematic review has synthesized the empirical evidence on cognitive apprenticeship and situative theory — the two pedagogical frameworks most often invoked to explain how novices become practitioners — within the undergraduate engineering capstone, using professional identity as the principal outcome. No review to date situates this synthesis against the specific obligation engineering graduates will be expected to discharge on entering practice: protecting the public’s safety, health, and welfare in decentralized technology contexts where the physical hazards are increasingly embedded (indoor air quality in schools, emissions from consumer-grade fabrication, cyber-physical failures in infrastructure, and unverified AI-assisted design decisions).

Filling this gap is both timely and actionable. For researchers, it consolidates a fragmented evidence base. For practitioners and program directors, it provides a theoretically coherent account of which capstone design features do identity work — and which do not. For the PFE and licensure communities (NSPE, NCEES, state boards), it offers the pedagogical warrant for treating capstone as the structural pivot between academic preparation and the public-facing responsibilities of licensed engineering practice — whether the next hazard takes the form of a particulate plume from a classroom 3D printer, an unverified AI-generated design, or a compromised building automation controller.

4 Theoretical Framework

The review uses a dual-framework lens:

Cognitive apprenticeship (Collins et al., 1989; 1991; Collins & Kapur, 2014) makes expert thinking visible through six methods — modeling, coaching, scaffolding, articulation, reflection, exploration — organized across four dimensions (content, method, sequencing, sociology). It is particularly productive for unpacking how capstone advisors and industry sponsors model the professional judgment that distinguishes “builder” from “engineer.”

Situative theory (Lave & Wenger, 1991; Wenger, 1998; Greeno, 1998; Johri & Olds, 2011) reframes learning as changing participation in a community of practice. It foregrounds legitimate peripheral participation, trajectories of participation, boundary crossing, and the co-construction of identity through practice. It is well-suited to capstone’s hybrid community — students, faculty advisors, industry sponsors, TAs, peers — where identity is negotiated across overlapping practices.

The two frameworks are complementary, not competing: cognitive apprenticeship specifies the pedagogical moves, situative theory specifies the social architecture in which those moves do their work. Treating them together is itself a contribution, because existing reviews address one or the other.

Professional identity is operationalized following Stevens, O’Connor, Garrison, Jocuns, & Amos (2008) and the broader PFE framework, attending to disciplinary affiliation, role adoption, professional values, and projected trajectory (including licensure intent, sector entry, and verification responsibility). The review will give particular attention to how studies in the corpus handle the epistemic dimension of professional identity — what counts as adequate evidence, who is authorized to verify an engineering artifact, what the professional’s responsibility is when the technology’s consequences are delayed, diffuse, or population-scale. This dimension is central to the Dartmouth and UF instrument development work (the EOTAI and PI-AVR scales under development for the planned NSF RFE mechanism study), and it is where most existing engineering-identity measures, including widely-used instruments from Godwin (2016) and Capobianco et al. (2012), do not currently reach.

5 Methods — Following PRISMA 2020

This review will follow the PRISMA 2020 Statement (Page, McKenzie, Bossuyt, et al., 2021) for reporting. A PRISMA-P protocol will be prepared and registered at the Open Science Framework (OSF), given that PROSPERO’s scope is formally restricted to health-related reviews. Registration will precede the formal search.

5.1 Eligibility Criteria

5.1.1 Inclusion

  • Empirical study (quantitative, qualitative, or mixed-methods)
  • Undergraduate engineering students
  • Setting is an engineering capstone design course, senior design, final-year project, or culminating design experience of at least one academic term
  • Study explicitly invokes cognitive apprenticeship and/or situative learning / situated learning / community of practice / legitimate peripheral participation as its theoretical framework (named in the framework, methods, or discussion)
  • Reports on professional identity, engineering identity, professional formation, professional socialization, or a clearly defined identity-adjacent construct (role identity, disciplinary affiliation, future-self projection, practitioner identity)
  • Peer-reviewed journal article or peer-reviewed conference proceedings (ASEE, FIE, REES, CDIO Proceedings, CEEA, SEFI)
  • Published in English
  • Published 1991 (publication year of Situated Learning, Lave & Wenger) through the search execution date
  • Full text obtainable through Dartmouth Library access or ILL

5.1.2 Exclusion

  • Theoretical or conceptual papers with no empirical data (collected separately for the discussion; not included in synthesis)
  • Editorials, opinion pieces, book reviews, dissertations and theses (dissertations used for snowballing only)
  • Studies of K–12 or graduate settings
  • Non-engineering disciplines (pre-med, business capstones, etc.), unless engineering is a substantive subpopulation with separable data
  • Studies of non-capstone design courses (cornerstone, sophomore-year), unless they include an explicit capstone comparison or follow-up
  • Studies in which the theoretical framework is named but not used (i.e., invoked only in background literature, with no analytic or design use)
  • Studies where identity is not an outcome or analytic category

A calibration set of 15 papers will be screened jointly by Emily Monroe and Eden Gray (and piloted with Jentry Campbell) to refine these criteria before full screening begins; disagreements will be logged and criteria revised if needed.

5.2 Information Sources

Planned databases, finalized with Jentry:

Database Coverage Rationale
ERIC (EBSCO) Core education literature, thesaurus terms
Scopus Interdisciplinary breadth, ASEE conference indexing
Web of Science Core Collection Citation tracking, high-impact journals
Engineering Village (Compendex + Inspec) Engineering-specific coverage
IEEE Xplore FIE conference proceedings, identity studies in CS/ECE
PsycINFO (APA) Identity development literature, psychometric instruments
ASEE PEER Primary repository for ASEE conference papers
ProQuest Education Database Additional education journal coverage

Supplementary sources: targeted hand-searching of Journal of Engineering Education, International Journal of Engineering Education, European Journal of Engineering Education, Advances in Engineering Education, Engineering Education Review, Studies in Engineering Education, and IEEE Transactions on Education; forward and backward citation chasing on included studies; contact of corresponding authors for ambiguous or in-press work.

Grey literature: Google Scholar (first 200 results per primary query); NSF-funded PFE project outputs via NSF PAR.

5.3 Search Strategy

Four concept blocks combined with Boolean AND; within each block, terms combined with OR. Controlled vocabulary (MeSH/thesaurus/subject headings) added per database by Jentry. Truncation (*), proximity operators (NEAR/n), and phrase quotes adapted per database syntax.

5.3.1 Concept Block 1 — Pedagogical / Theoretical Framework

"cognitive apprenticeship" OR "situated learning" OR "situative"
OR "situativity" OR "community of practice" OR "communities of practice"
OR "legitimate peripheral participation" OR "authentic learning"
OR "apprenticeship learning" OR "sociocultural learning"
OR (modeling AND coaching AND scaffolding)
OR (articulation AND reflection AND exploration)

5.3.2 Concept Block 2 — Capstone Context

capstone OR "senior design" OR "senior project" OR "final year project"
OR "final-year project" OR "culminating design" OR "culminating experience"
OR "design course" OR "design studio" OR "industry-sponsored project"
OR "client-based project" OR "integrated design" OR "integrative project"

5.3.3 Concept Block 3 — Discipline

engineer* OR "engineering education" OR "engineering design"
OR "mechanical engineering" OR "civil engineering" OR "electrical engineering"
OR "chemical engineering" OR "biomedical engineering" OR "industrial engineering"
OR "environmental engineering" OR "computer engineering" OR "software engineering"
OR "systems engineering" OR "aerospace engineering" OR "materials engineering"

5.3.4 Concept Block 4 — Outcome / Identity

"professional identity" OR "engineering identity" OR "professional formation"
OR "professional development" OR "professional socialization"
OR "role identity" OR "identity development" OR "identity formation"
OR "identity trajectory" OR "identification with" OR "becoming an engineer"
OR "practitioner identity" OR "sense of self as engineer"

5.3.5 Combination

(Block 1) AND (Block 2) AND (Block 3) AND (Block 4)

5.3.6 Contextual Sensitivity Search (non-restrictive)

Run as a separate context search to locate discussion-level literature; does not narrow the primary search, to avoid under-capturing identity work that predates the AI / public-health / cyber moment:

"artificial intelligence" OR "generative AI" OR "AI verification"
OR "human-in-the-loop" OR cybersecurity OR "cyber resilience"
OR "public health" OR "environmental health" OR "indoor air quality"
OR "occupational exposure" OR "safety engineering"
OR "professional licensure" OR "PE license" OR "FE exam" OR NCEES
OR "licensed engineer" OR "engineer in training" OR "EIT"

5.3.7 Database-Specific Variants

Full set of database-specific syntaxes appended to the final paper as supplementary material, per EER transparency expectations and PRISMA-S (Rethlefsen et al., 2021) reporting standards.

5.4 Study Selection Process

  1. De-duplication via Zotero import, then revtools package in R for fuzzy-matching title/author/DOI duplicates missed by exact-match de-duplication (typically 5–15% additional duplicates caught).
  2. Title/abstract screening in Covidence (available through Dartmouth Library subscription — see https://researchguides.dartmouth.edu/c.php?g=1519770). Dual independent screening by Emily Monroe (first author) and Eden Gray (second reviewer, summer 2026). Cohen’s κ reported; disagreements resolved by discussion; third-reviewer tie-breaker if needed (proposed escalation: Jentry Campbell → Dr. Melissa Richards, Dr. Sol Diamond, or Dr. Jennifer Love).
  3. Full-text screening with pre-registered reason codes for exclusion (e.g., no empirical data, not capstone, framework not operationalized, no identity outcome).
  4. Reference and citation chasing on included studies (Web of Science forward citations; backward reference-list scan).
  5. PRISMA 2020 flow diagram generated from Covidence export, using the PRISMA2020 R package (Haddaway et al., 2022), consistent with the existing Quarto/R workflow.

5.5 Data Extraction

Structured extraction form, piloted on five studies, capturing:

  • Bibliographic data (author, year, venue, country, discipline)
  • Study design and methodology (quant / qual / mixed; sample size; data sources)
  • Capstone structural features (duration, team size, sponsor type, advisor model, discipline, single- vs. multi-institution)
  • Theoretical framework deployment (cognitive apprenticeship only / situative only / both; which dimensions used; how applied — in design, analysis, or interpretation)
  • Identity construct and measurement (definition used; instrument or analytic approach; validity evidence)
  • Reported findings relating pedagogy to identity (effect sizes, themes, paraphrased representative quotes)
  • Treatment of transition to practice, licensure, public health outcomes, AI verification, or cyber-physical stewardship (yes/no; character of treatment)
  • Limitations and transferability claims

Extraction dual-coded for the first 20% of included studies; agreement reported; remaining studies single-coded with 20% random verification.

5.6 Risk of Bias / Methodological Quality Appraisal

The Mixed Methods Appraisal Tool (MMAT; Hong et al., 2018) will be used. MMAT accommodates five study types (qualitative, RCT, non-randomized, quantitative descriptive, mixed-methods) through parallel criteria sets, and is the field-standard tool for evidence syntheses spanning methodological traditions.

Appraisals conducted by two reviewers independently; discrepancies reconciled by discussion. Appraisal scores inform the strength of evidence language in the synthesis but are not used as exclusion criteria.

5.7 Synthesis

Given anticipated heterogeneity, meta-analysis is not planned. The synthesis uses convergent integrated synthesis (Lizarondo et al., 2022): quantitative findings are converted to qualitative statements (“qualitized”) and then thematically integrated with qualitative findings.

Three layers:

  1. Descriptive mapping — publication trends, geographies, disciplines, methodological distribution, theoretical framework distribution. Produced using bibliometrix on the included-set bibliographic metadata.
  2. Thematic synthesis (Thomas & Harden, 2008) organized around: (a) which cognitive apprenticeship methods are active in capstone; (b) which situative constructs are active in capstone; (c) how identity is conceptualized and measured; (d) how pedagogy and identity are linked in the reported evidence.
  3. Conceptual synthesis — integration into a framework positioning capstone as an identity-forming boundary practice between academic preparation and engineering work-in-the-world, with specific attention to public-health-facing, AI-verification, and cyber-physical stewardship responsibilities.

6 Analysis Tooling Stack (R / Quarto)

All analysis runs in R via Quarto, consistent with the existing workflow. A project-level renv lockfile will be committed alongside the protocol for reproducibility. The tooling stack below is ordered roughly in the order each piece will be used.

6.2 ML-Assisted Screening via ASReview (optional)

If the Covidence queue exceeds ~2,000 records after deduplication, ASReview reorders screening so that likely-includes appear first and Eden and Emily can stop screening when marginal yield drops below a pre-registered threshold. Published validation shows 70–90% workload reduction without losing relevant records. Decision point: if database hits exceed 2,000 post-dedup, adopt ASReview; otherwise screen all records in Covidence directly. Whichever path is taken will be reported in the manuscript per PRISMA-S transparency requirements.

6.3 Record-Keeping Structure

Reproducibility requires a single source of truth for every artifact the review touches. The proposed project directory lives on encrypted Dartmouth OneDrive, mirrored to the OSF project for registration and archival, with version control via git for all code and protocol documents.

cogapp-situative-capstone-SR/
├── README.md                          # Entry point, directory map, contact info
├── renv.lock                          # Locked R package versions
├── _quarto.yml                        # Project-wide render settings
├── protocol/
│   ├── protocol_v0.2.qmd              # This document
│   ├── protocol_v0.2.pdf              # Rendered version for circulation
│   ├── osf_registration.md            # OSF registration text
│   └── changelog.md                   # Version history with dated entries
├── search/
│   ├── search_strings/
│   │   ├── master_strategy.md         # Block-level master (Section 5.3)
│   │   ├── scopus.txt                 # Scopus-specific syntax
│   │   ├── wos.txt                    # Web of Science syntax
│   │   ├── eric.txt                   # ERIC syntax
│   │   └── ...                        # one file per database
│   ├── database_exports/
│   │   ├── scopus_YYYY-MM-DD.ris
│   │   ├── wos_YYYY-MM-DD.ris
│   │   └── ...                        # ISO-dated raw exports, never edited
│   ├── mock_exports/                  # Simulated corpus for pipeline testing
│   └── search_log.md                  # Dated entries: who ran what, when, count
├── screening/
│   ├── covidence_project_link.md      # URL + access notes
│   ├── calibration_set.csv            # The 15 training papers
│   ├── calibration_decisions.csv      # Both reviewers' calls + reconciliation
│   ├── screening_decisions.csv        # Final title/abstract decisions
│   ├── fulltext_decisions.csv         # Full-text decisions + reason codes
│   ├── cohens_kappa.R                 # IRR computation
│   └── reconciliation_log.md          # Every disagreement, dated, resolved
├── extraction/
│   ├── extraction_template.xlsx       # Piloted extraction form
│   ├── extracted_data.csv             # Long-format tidy extraction
│   ├── mmat_appraisal.csv             # MMAT scores per study
│   └── extraction_notes/              # Per-study free-text notes
├── analysis/
│   ├── 01_deduplication.R             # revtools fuzzy dedup
│   ├── 02_prisma_flow.qmd             # PRISMA2020 rendering
│   ├── 03_bibliometric_mapping.qmd    # Descriptive map
│   ├── 04_thematic_synthesis.qmd      # Thomas & Harden coding
│   ├── 05_conceptual_synthesis.qmd    # Integration
│   └── figures/                       # Rendered outputs
├── manuscript/
│   ├── manuscript.qmd                 # EER submission draft
│   ├── supplementary_material.qmd     # Full search strings, extraction tables
│   ├── references.bib                 # Master bib, citation keys stable
│   └── cover_letter.md
├── team/
│   ├── roles_credit.md                # CRediT assignments
│   ├── meeting_notes/                 # Dated .md files per meeting
│   └── onboarding/
│       ├── eden_onboarding.md         # Day-1 packet
│       ├── calibration_training.md    # Screening training protocol
│       └── reading_list.md            # Foundational references for Eden
└── .gitignore

Four operating rules keep this useful instead of ornamental:

  1. Raw is raw. Database exports go into search/database_exports/ with an ISO date in the filename and are never edited. Any cleaning happens in a script that reads the raw file and writes a cleaned file to screening/. If a reviewer needs to re-run something from scratch, the raw export is the ground truth.
  2. Every decision is dated and attributable. Screening decisions carry reviewer initials and a timestamp; the reconciliation log records every disagreement and how it resolved. This is the record that makes PRISMA-S reporting and OSF registration credible, and it’s also the record that gets Eden legitimate co-authorship credit.
  3. Scripts, not spreadsheets, for anything that needs to run twice. Deduplication, IRR computation, PRISMA flow, bibliometric mapping — these live in R scripts under analysis/, not in Excel. Spreadsheets are fine for human-readable extraction tables; they are not fine for anything that will need to be reproduced.
  4. Weekly git commits minimum; daily during active screening. Commit messages describe what changed and why, not just what file. “Updated screening_decisions.csv” is useless; “Reconciled 12 disagreements from week 2 calibration; 3 criteria clarifications logged” is the record future-Emily and peer reviewers will actually need.

The OSF project is set up now (before the search runs) with the registered protocol, the directory README, and the search strategy locked. Once the search runs, raw exports and de-duplicated sets are uploaded. Once screening concludes, the screening logs and reconciliation log are uploaded. Data availability statement in the EER manuscript points to the OSF DOI.

7 Connection to the Dissertation

This review is the theoretical grounding for the dissertation’s capstone intervention chapter and directly underwrites three downstream empirical strands:

  • The NCEES Foundation–funded capstone intervention study (“Licensed Engineer as Human-in-the-Loop: A Capstone Framework for Developing Safety-Conscious Engineering Leaders in an AI-Centric Future”): the review establishes the theoretical warrant, identifies how existing engineering-identity instruments (Godwin, 2016; Capobianco et al., 2012) do and do not capture the verification and public-responsibility dimensions, and specifies the measurement gap that the EOTAI and PI-AVR scales under development are designed to fill.
  • The planned NSF RFE three-aim mechanism study (Dartmouth + UF): the review’s conceptual framework becomes the theoretical grounding for the hypothesized mediated pathway (cognitive apprenticeship experiences → epistemic practice shift → professional identity formation), and identifies the construct landscape into which the EOTAI and PI-AVR scales must fit.
  • The “Safer by Design” thread connecting SHIELD-3D, the makerspace IAQ work, and the cyber-resiliency research: the review articulates the pedagogical mechanism by which identity-formed engineers become the human-in-the-loop safeguard on which safer systems depend — the professional judgment that says “this ventilation is not adequate,” “this AI-assisted design has not been verified,” or “this firmware has not been security-reviewed.”

8 Alignment with Engineering Education Review

EER Requirement Planned Compliance
Review article length Target 6,000–7,500 words main text (> 5,000 minimum)
Abstract ≤ 400 words, structured: Background / Methods / Findings / Implications
References 50–80 expected; APA alphabetical per EER style
Language American English; consistent spelling and terminology
Review type disclosure Stated as systematic review using PRISMA 2020 and PRISMA-S reporting standards; MMAT appraisal; convergent integrated synthesis
Inclusion criteria, search strategy, analytic methods Detailed in Methods; full database-specific syntaxes in Supplementary Material
Theoretical contribution Integrated conceptual framework for capstone as a PFE-anchored, identity-forming boundary practice under public-health, AI, and cyber-physical conditions
New directions Explicit research agenda, including recommended instrumentation (anchoring the EOTAI and PI-AVR development work) and study designs
GenAI disclosure Transparent statement per EER policy: tools, version, use, and review process
Data availability Search strategies, screening logs, extraction tables deposited on OSF with DOI
Ethics Secondary literature review; no human subjects; stated explicitly

9 Team Structure and CRediT Roles

  • Emily Monroe (first author): Conceptualization, Methodology, Investigation, Formal analysis, Writing – Original Draft, Writing – Review & Editing, Project administration, Funding acquisition.
  • Jentry Campbell (co-author, proposed): Methodology (information retrieval), Data Curation (search documentation), Writing – Review & Editing.
  • Eden Gray (co-author, proposed): Investigation (dual independent screening, extraction), Data Curation (screening logs, extraction tables, PRISMA flow diagram), Writing – Review & Editing.
  • Second reviewer backup (if Eden unavailable): TBD Thayer PhD peer.
  • Senior mentor (TBD — Melissa Richardson, Sol Diamond or Jennifer Love): Supervision, tie-breaker on unresolved screening or extraction disagreements, Writing – Review & Editing.

Co-authorship for Eden is justified on substantive grounds: she is the second independent reviewer on screening (a PRISMA-mandated role), a co-extractor, and a co-author on the PRISMA flow diagram and bibliometric mapping outputs. This is a legitimate undergraduate co-authorship on a peer-reviewed systematic review — an unusually strong CV line for graduate or industry applications, and appropriate recognition for the level of contribution required.

10 Schedule

Total duration: ~11 months from protocol finalization to submission, structured around Dartmouth Summer Term 2026 (classes begin June 25, 2026; classes end August 26, 2026; final examinations end September 1, 2026). Eden’s 10-week period is scheduled against the summer term calendar with the assumption that she is concurrently enrolled in coursework — effort is therefore scoped at ~20 hours/week, not at a full-time 30 hours/week. If Eden is not enrolled in classes during her 10 weeks, hours can scale up and the timeline can compress accordingly.

10.1 Phase 1 — Late April through late June 2026 (pre-Eden preparation)

Week of Lead Deliverable
Apr 27 Emily Protocol v0.2 finalized; first meeting with Jentry to resolve Section 14 open questions
May 4 Emily + Jentry Revisions from Jentry meeting; concept blocks refined; database-specific syntaxes drafted
May 11 Emily + Jentry OSF registration submitted; DOI assigned; search strategy locked
May 18 Emily Pipeline simulation (mock corpus → revtools dedup → PRISMA render); directory structure committed to git; Eden’s Covidence seat requested via Dartmouth Library
May 25 – Jun 7 Jentry Search execution across 8 databases; raw exports dated and committed to git
Jun 8 Emily + Jentry De-duplication (Zotero exact match + revtools fuzzy match); Covidence import; record-level counts locked for PRISMA Identification stage
Jun 15 Emily Calibration set of 15 papers pre-selected from deduplicated corpus; Eden onboarding packet finalized; reading list shared with Eden one week before her start date
Jun 22 Emily Pre-arrival slack: buffer for anything that ran long; confirm Dartmouth Library systematic-review workshop time with Jentry for Eden’s Week 1

10.2 Phase 2 — June 25 through September 1, 2026 (Eden’s 10 weeks)

Mapped to the 2026 Dartmouth Summer Term calendar, accounting for the July 3 Independence Day holiday (no classes) and the August 27–28 pre-examination break + August 29–September 1 final examinations. Eden’s Week 10 coincides with pre-exam break and finals, so it is scoped as a lighter wrap-up week — contemplative work (Methods-section drafting, handoff document) that fits around exam preparation.

Eden Week Calendar Activity Eden’s Hours Emily’s Hours
1 Jun 25 – Jul 1 (Thu start) Orientation; read foundational references (Lave & Wenger, Collins et al., Johri & Olds); read protocol; Covidence training; Dartmouth Library systematic-review workshop with Jentry; OSF project orientation 15 6
2 Jul 2 – Jul 8 (Jul 3 holiday) Calibration set (15 papers) dual-screened independently; reconciliation meeting; criteria refinement documented and committed to git 15 8
3 Jul 9 – Jul 15 Title/abstract screening begins in Covidence, dual independent 22 10
4 Jul 16 – Jul 22 Title/abstract screening continues; mid-phase IRR check (Cohen’s κ) on first 200 screened; reconciliation of outstanding disagreements 22 12
5 Jul 23 – Jul 29 Title/abstract screening continues; full-text retrieval initiated (Eden handles ILL requests in parallel) 22 10
6 Jul 30 – Aug 5 Title/abstract screening completed; PRISMA Screening-stage counts locked; full-text screening begins 22 12
7 Aug 6 – Aug 12 Full-text screening, dual independent; reconciliation on full-text disagreements; PRISMA Eligibility-stage counts locked 22 12
8 Aug 13 – Aug 19 Full-text screening completed; structural-features extraction (Eden lead) begins on confirmed includes 22 12
9 Aug 20 – Aug 26 (classes end Aug 26) Extraction of theoretical-framework usage (Emily lead, Eden secondary) on included set; MMAT appraisal begins 20 15
10 Aug 27 – Sep 1 (pre-exam break + finals) Lighter week: Eden drafts the Methods section of the manuscript (Search, Screening, Extraction subsections) — work that fits around final exams; handoff document for post-summer work; final PRISMA flow diagram rendered 8 8
Total ~190 ~105

10.3 Phase 3 — September 2026 – March 2027 (post-summer synthesis and drafting)

Month Lead Activity
September Emily Bibliometric descriptive mapping; thematic synthesis begins
October Emily Thematic synthesis completed; conceptual synthesis begins
November Emily Full draft assembled; Eden and Jentry review Methods; Sol/Jennifer review conceptual framework
December Emily Revisions; GenAI disclosure finalized; supplementary materials packaged
January All Internal review cycle; address all co-author comments
February Emily Final pre-submission polish; OSF supplementary files updated
March Emily EER submission

10.4 Phase 4 — Post-submission (variable)

Peer review response, revised manuscript preparation, search top-up if >6 months between submission and acceptance. Emily leads the communication with journal editors. All co-authors remain engaged through acceptance.

11 Eden’s Onboarding — What Day 1 Looks Like

To make Eden’s summer productive from the start, the onboarding packet (team/onboarding/eden_onboarding.md) includes:

  1. Context briefing (written, ~2,000 words): the dissertation theme, the NCEES grant, why this review exists, who Jentry is, what the NSF RFE submission depends on, what the EOTAI and PI-AVR scales are for.
  2. Foundational reading list, in this order: Lave & Wenger (1991) chapters 1–3; Collins, Brown, & Holum (1991); Johri & Olds (2011); Pierrakos et al. (2024); Stevens et al. (2008); Morelock (2017). A half-page reflection on each, not for grading but for Eden’s orientation and for surfacing any conceptual misalignments before screening begins.
  3. Methods training: PRISMA 2020 primer; MMAT walk-through with one pre-worked example; Covidence tutorial; Dartmouth Library systematic-review workshop (scheduled with Jentry in week 1).
  4. Tools setup: OneDrive access, OSF account, Covidence seat, Zotero group library, GitHub access to the project repository, R + RStudio on her laptop with the renv lockfile synced.
  5. Expectations on co-authorship: the CRediT roles above; expected contribution to the Methods section draft; acknowledgment that her name appears on the registered protocol and on any conference presentations; commitment to remain available for post-summer email review.
  6. Communication cadence: weekly 30-minute 1:1s Monday mornings; daily async check-ins via shared channel; calendar for group meetings with Jentry (biweekly) and senior mentor (monthly).
  7. Stipend, hours, and scope boundaries: what counts as “in scope” for the ~190 hours, what doesn’t, how to flag scope creep, how to request help when stuck. If Eden’s hours actually scale up (not taking classes concurrently), extraction can expand to include a first-pass thematic coding sweep.

12 Draft Research Statement (for OSF Registration and EER Manuscript)

Undergraduate engineering capstone design courses occupy a structurally singular position in the preparation of engineers: they are simultaneously the culmination of academic preparation and the first authorized approximation of professional engineering work. Two learning theories — cognitive apprenticeship (Collins, Brown, & Newman, 1989) and situative learning theory (Lave & Wenger, 1991) — have each been invoked to explain how capstone does this boundary work, but no synthesis has examined them together, at capstone specifically, with professional identity as the outcome of interest. This gap matters because the engineering workforce that capstone students enter carries public-facing responsibilities that are widening rather than narrowing: environmental and public-health exposures embedded in consumer-grade fabrication and building systems, AI-assisted design decisions that demand human-in-the-loop verification, and cyber-physical vulnerabilities that cross every sub-discipline. All three are failure modes of engineered systems that professional judgment is meant to prevent, and all three require an engineer whose identity has been formed — not merely trained. This systematic review synthesizes peer-reviewed empirical studies (1991–2026) that operationalize cognitive apprenticeship and/or situative theory within engineering capstone courses and report on professional identity formation, following PRISMA 2020 and PRISMA-S standards, using the MMAT for quality appraisal and convergent integrated synthesis for analysis. The review will produce (a) a descriptive map of the evidence base; (b) a thematic account of which pedagogical mediators do identity work in capstone and under what conditions; (c) a critical assessment of how identity has been measured, including a diagnosis of where existing instruments (e.g., Godwin, 2016; Capobianco et al., 2012) do and do not reach the verification and public-responsibility dimensions of contemporary engineering practice; and (d) an integrated conceptual framework positioning capstone as an identity-forming boundary practice between academic preparation and engineering work-in-the-world — with specific implications for the Professional Formation of Engineers agenda, for PE licensure pathways, and for preparing graduates to serve as the professional safeguard on which safer systems for humans depend.12. Foundational References (Seed Set)

This list is the theoretical and methodological backbone. It is not the review’s final reference list — that is produced by the review itself — but it anchors the framing and will recur in the discussion.

Theoretical foundations

Brown, J. S., Collins, A., & Duguid, P. (1989). Situated cognition and the culture of learning. Educational Researcher, 18(1), 32–42.

Collins, A., Brown, J. S., & Newman, S. E. (1989). Cognitive apprenticeship: Teaching the crafts of reading, writing, and mathematics. In L. B. Resnick (Ed.), Knowing, learning, and instruction (pp. 453–494). Lawrence Erlbaum.

Collins, A., Brown, J. S., & Holum, A. (1991). Cognitive apprenticeship: Making thinking visible. American Educator, 15(3), 6–11, 38–46.

Collins, A., & Kapur, M. (2014). Cognitive apprenticeship. In R. K. Sawyer (Ed.), The Cambridge handbook of the learning sciences (2nd ed., pp. 109–127). Cambridge University Press.

Greeno, J. G. (1998). The situativity of knowing, learning, and research. American Psychologist, 53(1), 5–26.

Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge University Press.

Wenger, E. (1998). Communities of practice: Learning, meaning, and identity. Cambridge University Press.

Engineering education theoretical applications

Johri, A., & Olds, B. M. (2011). Situated engineering learning: Bridging engineering education research and the learning sciences. Journal of Engineering Education, 100(1), 151–185.

Stevens, R., O’Connor, K., Garrison, L., Jocuns, A., & Amos, D. M. (2008). Becoming an engineer: Toward a three dimensional view of engineering learning. Journal of Engineering Education, 97(3), 355–368.

Identity measurement in engineering (construct landscape for EOTAI/PI-AVR positioning)

Capobianco, B. M., French, B. F., & Diefes-Dux, H. A. (2012). Engineering identity development among pre-adolescent learners. Journal of Engineering Education, 101(4), 698–716.

Godwin, A. (2016). The development of a measure of engineering identity. ASEE Annual Conference Proceedings, Paper ID #14814.

Adjacent systematic reviews (gap evidence)

Donaldson, T., Fore, G. A., Filippelli, G. M., & Hess, J. L. (2020). A systematic review of the literature on situated learning in the geosciences: Beyond the classroom. International Journal of Science Education, 42(5), 722–743. https://doi.org/10.1080/09500693.2020.1727060

Minshew, L. M., Olsen, A. A., & McLaughlin, J. E. (2021). Cognitive apprenticeship in STEM graduate education: A qualitative review of the literature. AERA Open, 7.

Morelock, J. R. (2017). A systematic literature review of engineering identity: Definitions, factors, and interventions affecting development, and means of measurement. European Journal of Engineering Education, 42(6), 1240–1262.

Pierrakos, O., Pappas, J., Crowe, W., Di Vittorio, C., Holcomb, T., & Luthy, K. (2024). Educating the whole engineer at Wake Forest Engineering: Using cognitive apprenticeship as an effective pedagogical approach to cultivate design learning, team effectiveness, entrepreneurial mindset and character in capstone design. International Journal of Engineering Education, 40(6), 1507–1530.

Tenhunen, S., Männistö, T., Luukkainen, M., & Ihantola, P. (2023). A systematic literature review of capstone courses in software engineering. Information and Software Technology, 159, 107191.

Methodological standards

Haddaway, N. R., Page, M. J., Pritchard, C. C., & McGuinness, L. A. (2022). PRISMA2020: An R package and Shiny app for producing PRISMA 2020–compliant flow diagrams. Campbell Systematic Reviews, 18(2).

Hong, Q. N., Pluye, P., Fàbregues, S., Bartlett, G., Boardman, F., Cargo, M., et al. (2018). Mixed Methods Appraisal Tool (MMAT), version 2018. Canadian Intellectual Property Office, Registration No. 1148552.

Lizarondo, L., Stern, C., Carrier, J., Godfrey, C., Rieger, K., Salmond, S., Apóstolo, J., Kirkpatrick, P., & Loveday, H. (2022). Chapter 8: Mixed methods systematic reviews. In E. Aromataris & Z. Munn (Eds.), JBI Manual for Evidence Synthesis. JBI.

Page, M. J., McKenzie, J. E., Bossuyt, P. M., et al. (2021). The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 372, n71.

Rethlefsen, M. L., Kirtley, S., Waffenschmidt, S., Ayala, A. P., Moher, D., Page, M. J., Koffel, J. B., & PRISMA-S Group. (2021). PRISMA-S: An extension to the PRISMA Statement for Reporting Literature Searches in Systematic Reviews. Systematic Reviews, 10(1), 39.

Thomas, J., & Harden, A. (2008). Methods for the thematic synthesis of qualitative research in systematic reviews. BMC Medical Research Methodology, 8, 45.

13 Open Questions to Resolve with Jentry at the First Meeting

  1. OSF registration — confirm that OSF is the right registry (PROSPERO’s scope is health-restricted) and set up the project before the search runs.
  2. Scope of “engineering” — include engineering technology programs, or restrict to ABET EAC-accredited engineering? Recommendation: include both, with a planned subgroup analysis.
  3. Conference proceedings — ASEE and FIE are the dominant venues for capstone identity work; excluding them would under-represent the field. Recommendation: include peer-reviewed conference proceedings.
  4. Covidence vs. Rayyan — Dartmouth Library provides Covidence access; confirm seat availability for Emily + Eden + Jentry.
  5. Extraction tool — custom Quarto/R workflow (matches the existing pipeline) vs. Covidence extraction forms. Recommendation: Covidence for structured fields + R for interpretive framework coding.
  6. Search update cadence — single search with top-up before resubmission of any revised manuscript, per EER’s typical review cycle.
  7. Eden’s Dartmouth Library training slot — schedule in Week 1 of Eden’s summer term.

End of protocol draft v0.2. Next step: walk this document through with Jentry, finalize the four concept blocks and database-specific syntaxes, register the protocol on OSF, lock the search, run the pipeline simulation before Eden arrives.