# Designing and Conducting Mixed Methods Research — Work the Book **Assessment:** Creswell and Plano Clark give you a usable design discipline: name what combining two forms of evidence must accomplish, then show the procedure and inference that accomplish it. Their strongest contribution is the trace from research question to integration. Claims that combining methods gives a better understanding need conditions; a second strand can repeat a bias, miss the same people, or compare a different construct. The appendices demonstrate procedures and acknowledge limits, including excluded questionnaire domains and unmeasured health outcomes. This assessment covers selected paragraphs, not the entire textbook. # Prompt 0 — Retrieval and vault setup The inventory alias **Creswell Clark Plano Mixed Methods** resolves to the title and two authors above. Native title page U005 names both authors and Third Edition; copyright page U006 names SAGE, paperback ISBN 9781483344379 and copyright 2018. OPF gives 2017-08-30; the publisher lists August 2017. This note uses publication year 2017 and retains the copyright distinction. It does not infer an ebook ISBN from paperback metadata. [Publisher record](https://www.sagepub.com/shop/buy-a-book/designing-and-conducting-mixed-methods-research-3-241842). The original local Drive path is unavailable. A metadata-grounded raw connector fetch supplies the complete 8,171,699-byte EPUB. SHA-256 is recorded above; all 366 ZIP entries pass CRC, and all 256 spine documents pass strict XML. Integrity proves bytes and parseability, not a full reading or reproduction of the book’s research. One bounded Readwise title/semantic search requests three matches. The connector returns six unrelated highlights from Rick Warren and the computing-education handbook; none matches this title or edition. No reader highlights, dates, colors, notes or ratings are imported. The supplied EPUB replaces Readwise as Layer 1. Every selection and emphasis below belongs to the system. No claim about an unread section’s absence is permitted. No existing canonical review matches the native title/authors in the review-folder search. Existing research notes supply adjacency. User instructions keep source excerpts, literature notes, permanent-note drafts and visual maps inside this one file; no separate highlights, visual, hub or permanent-note files are created. No reciprocal edits to other notes occur. Amazon search yields no title-matched result and the paperback destination returns an HTML shell without a usable identity match; the Amazon field stays null. ## Coverage and location contract 20,425/188,836 extracted p/h1–h4/li block words (10.8%); 50 of 256 spine items sampled; no whole chapter read in full; tables, image figures, standalone blockquotes, index and bibliography not evaluated The excerpt location `CPC-U###-e###` means one-based spine order and zero-based index into the native XHTML `body.iter()` tree. It does not mean page, Kindle location or filtered paragraph number. Native file mapping and exact read-element lists follow. Paragraph extraction includes p, h1–h4 and li elements; nested lists may duplicate text, so the word counts describe extracted blocks, not unique book words. No whole chapter is marked full. Native standalone blockquotes, tables, image figures and the bibliography are not assessed. Apparent gaps in quoted definitions in the extraction are extraction limits, not missing definitions in the book. | Spine unit / native file | Read elements | Coverage | | --- | --- | --- | | U006 / `index_split_004.html` | e041, e048, e050, e052, e054, e058 | Selected blocks only | | U019 / `index_split_017.html` | e001, e004, e006, e008, e018, e022, e023, e026, e028, e029, e032, e036, e037, e038, e039, e040 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U021 / `index_split_019.html` | e001, e004, e005, e006 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U022 / `index_split_020.html` | e001, e004, e005 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U029 / `index_split_027.html` | e010, e015 | Selected blocks only | | U031 / `index_split_029.html` | e001, e004, e005, e006, e008 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U032 / `index_split_030.html` | e001, e004, e007, e008, e009, e010, e012, e013 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U033 / `index_split_031.html` | e004 | Selected blocks only | | U038 / `index_split_036.html` | e004 | Selected blocks only | | U043 / `index_split_041.html` | e004, e008 | Selected blocks only | | U044 / `index_split_042.html` | e004 | Selected blocks only | | U050 / `index_split_048.html` | e004 | Selected blocks only | | U051 / `index_split_049.html` | e004, e005 | Selected blocks only | | U052 / `index_split_050.html` | e004 | Selected blocks only | | U055 / `index_split_053.html` | e004 | Selected blocks only | | U057 / `index_split_055.html` | e004, e029, e043, e063, e065, e067 | Selected blocks only | | U058 / `index_split_056.html` | e001, e004, e006, e009, e010, e013, e016, e017, e019, e020, e021, e022, e025, e027, e028, e031, e034, e042, e045, e046, e049, e052, e053, e054, e056, e058, e061, e064, e065, e066, e067, e069, e072, e075, e077, e079, e082, e084, e085, e087, e088, e089, e090 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U059 / `index_split_057.html` | e004, e038, e043, e069, e070, e072, e074, e077 | Selected blocks only | | U070 / `index_split_068.html` | e004, e005 | Selected blocks only | | U074 / `index_split_072.html` | e004, e005, e011, e031, e037, e068, e073, e077, e080, e081, e091 | Selected blocks only | | U088 / `index_split_086.html` | e004 | Selected blocks only | | U091 / `index_split_089.html` | e004, e023, e027, e038, e039, e040, e056, e057, e058, e060 | Selected blocks only | | U096 / `index_split_094.html` | e004 | Selected blocks only | | U101 / `index_split_099.html` | e001, e004, e007, e008, e009, e010, e012, e018, e023, e025 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U104 / `index_split_102.html` | e004 | Selected blocks only | | U105 / `index_split_103.html` | e004 | Selected blocks only | | U122 / `index_split_120.html` | e001, e004, e005, e006, e009, e011, e013 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U124 / `index_split_122.html` | e004 | Selected blocks only | | U125 / `index_split_123.html` | e004 | Selected blocks only | | U126 / `index_split_124.html` | e004 | Selected blocks only | | U128 / `index_split_126.html` | e001, e004, e011, e012, e013 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U129 / `index_split_127.html` | e001, e004 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U130 / `index_split_128.html` | e004 | Selected blocks only | | U143 / `index_split_141.html` | e001, e004, e005, e009, e010, e011, e012, e013, e016, e017, e018, e019, e020, e021, e022, e024, e025, e026, e028, e029, e033, e038, e039, e040, e041, e042, e043, e047, e049, e051, e053, e056, e059 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U148 / `index_split_146.html` | e004 | Selected blocks only | | U153 / `index_split_151.html` | e001, e004, e008 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U154 / `index_split_152.html` | e004 | Selected blocks only | | U160 / `index_split_158.html` | e004 | Selected blocks only | | U164 / `index_split_162.html` | e004, e005 | Selected blocks only | | U190 / `index_split_188.html` | e004 | Selected blocks only | | U191 / `index_split_189.html` | e004 | Selected blocks only | | U194 / `index_split_192.html` | e004 | Selected blocks only | | U195 / `index_split_193.html` | e004 | Selected blocks only | | U197 / `index_split_195.html` | e001, e004 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U199 / `index_split_197.html` | e001, e004 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U205 / `index_split_203.html` | e004 | Selected blocks only | | U223 / `index_split_221.html` | e004 | Selected blocks only | | U238 / `index_split_236.html` | e004 | Selected blocks only | | U244 / `index_split_242.html` | e001, e004, e007, e043, e052, e053, e058, e059, e062, e063, e066, e071, e074, e083, e085, e088 | All nonempty extracted p/h1–h4/li blocks; other elements excluded | | U246 / `index_split_244.html` | e092, e104 | Selected blocks only | All other spine units are not evaluated; U005 establishes identity and source metadata, while the structural map uses heading inventory and publisher contents. Unlisted elements in sampled units are not evaluated. ## Embedded Layer 1 source-selection index Sixteen exact native substrings appear below. Whitespace is normalized; punctuation and wording remain native. Adjacent selections from one textbook do not constitute independent confirmation. S-14–16 come from reproduced articles and retain their original attribution. | ID | Exact excerpt | Stable native location | Attribution | | --- | --- | --- | --- | | S-01 | “integrates (or mixes or combines) the two forms of data and their results,” | `CPC-U019-e037` | Creswell and Plano Clark: core characteristic | | S-02 | “Not all situations justify the use of mixed methods.” | `CPC-U021-e005` | Creswell and Plano Clark: limits | | S-03 | “The results from the quantitative and qualitative data may be contradictory,” | `CPC-U022-e005` | Creswell and Plano Clark: rationale | | S-04 | “We strongly recommend that researchers first gain experience with both quantitative research and qualitative research separately before undertaking a mixed methods study.” | `CPC-U031-e004` | Creswell and Plano Clark: prerequisites | | S-05 | “Are the skills and personnel available to complete this study?” | `CPC-U032-e009` | Creswell and Plano Clark: feasibility | | S-06 | “the use of mixed methods arises due to issues that develop during the process of conducting the research.” | `CPC-U050-e004` | Creswell and Plano Clark: emergent design | | S-07 | “Contradictions may provide new insights into the topic,” | `CPC-U057-e067` | Creswell and Plano Clark: divergence | | S-08 | “There are two points where integration occurs in an explanatory sequential design.” | `CPC-U058-e045` | Creswell and Plano Clark: integration | | S-09 | “the researcher needs to decide which results to use from the qualitative phase to build the quantitative feature” | `CPC-U059-e074` | Creswell and Plano Clark: selection | | S-10 | “which is not to suggest they are more rigorous than the core designs;” | `CPC-U070-e005` | Creswell and Plano Clark: complexity | | S-11 | “Use consistent terms to refer to variables/phenomena examined across multiple questions.” | `CPC-U091-e056` | Creswell and Plano Clark: question writing | | S-12 | “the integration intent, the integration data analysis procedures, the representation of the integration results, and the interpretation of the integration results.” | `CPC-U122-e005` | Creswell and Plano Clark: four considerations | | S-13 | “whether the researcher actually integrated the two methods as opposed to collecting data for both strands and keeping them separate throughout the study.” | `CPC-U143-e051` | Creswell and Plano Clark: evaluation | | S-14 | “it does not evaluate sexual violence, electronic media violence (using text messages, emails, etc.), or stalking.” | `CPC-U197-e004` | Enosh, Tzafrir, and Stolovy: Appendix C limitation; Q3P | | S-15 | “we sought input on the accuracy of our findings and recommendations for implementing changes” | `CPC-U244-e088` | Greysen and colleagues: Appendix F; Q3P | | S-16 | “we did not collect outcomes of the transitions in care our participants experienced, so we cannot describe the clinical impact of poorly-coordinated transitions.” | `CPC-U246-e104` | Greysen and colleagues: Appendix F limitation; Q3P | Selection clusters: definition and feasibility (S-01–05), design interfaces (S-06–10), explicit question/analysis/reporting (S-11–13), and empirical limits (S-14–16). Density follows system interest in integration and validity; it does not measure your interests. Timestamps and highlight colors are unavailable. # Prompt 1 — Setup, contract, baselines, biases and priors > [!info] System-inferred reading contract > No reader purpose, existing-knowledge statement or personal response is supplied. The proposed purpose is to evaluate the book as a research-design reference. Correct this contract before treating any suggested application as your commitment. You could use the book to choose between comparison, explanation and instrument development; then test whether a proposed study states an actual integration procedure. Success means you can identify the question, both sampling frames, the interface, a combined inference and one threat to that inference. Evidence that would change the system assessment includes comparative studies showing no added insight or an integrated inference that fails because the strands measure different things. This book cannot establish that every mixed methods study beats a single-method study, that a well-written report is valid, or that any reproduced intervention improves outcomes. Personal beliefs, motives, resistance, incentives and experience remain unknown. The system risks favoring the textbook’s typology because it is easy to organize and favoring joint displays because they are inspectable. Those biases would surface in Prompts 4, 7 and 13. Publication incentives are author-discussed: U029 e015 says one study can yield several publications. The risk is fragmenting a study before explaining its combined inference; no motive is assigned to you. ## Open questions | ID | Question | Category | | --- | --- | --- | | O01 | What distinguishes integration from two parallel reports? | Definition | | O02 | Does coding interview text into counts create a second research strand? | Definition | | O03 | How does an interface create a new inference? | Mechanism | | O04 | When can both strands repeat the same bias? | Evidence | | O05 | Which problem needs comparison instead of explanation? | Design | | O06 | What makes exploratory development culture-sensitive? | Transfer | | O07 | How do you select which initial results deserve follow-up? | Selection | | O08 | Who is absent from the initial sampling frame? | Inclusion | | O09 | How do different samples support one combined inference? | Sampling | | O10 | When is divergence a construct mismatch? | Failure mode | | O11 | What does a mixed methods question add beyond two subquestions? | Writing | | O12 | Can the interface be planned before an emergent phase? | Feasibility | | O13 | Who can contest the interpretation in participatory research? | Agency | | O14 | Does collaboration measure a change in health or policy outcomes? | Evidence | | O15 | What makes a joint display inferentially sound? | Analysis | | O16 | Does this taxonomy privilege a particular disciplinary tradition? | Context | | O17 | What validity threat arises only after combining strands? | Validity | | O18 | How do access needs affect the information a study receives? | Accessibility | | O19 | Which questionnaire dimensions are omitted because nobody reports them? | Silence | | O20 | Which software, reporting standards and publication advice need current checks? | Drift | ## Priors seed table These are system starting judgments, not reader beliefs and not numerical probabilities. Prompt 7 updates them only after inspecting the identified evidence. | Claim | Initial judgment | Confidence | Reason | Evidence needed | | --- | --- | --- | --- | --- | | C-001 | Integration needs a boundary | Provisional | A label must exclude cases | Native definition | | C-003 | Added information need not improve inference | Provisional | Mismatch can add noise | Cases and comparative evaluation | | C-005 | Phase dependencies constrain feasible work | Provisional | Later work waits for earlier results | Concrete feasibility guidance | | C-007 | Typologies can help and constrain | Provisional | Labels simplify adaptive procedures | Alternative approaches | | C-009 | Discordance can identify errors or new questions | Provisional | Several mechanisms fit disagreement | Reanalysis example | | C-010 | Explanatory accounts need a causal limit | Provisional | Participant accounts can omit causes | Procedure and inference distinction | | C-011 | Local themes can improve relevance but miss domains | Provisional | Silence affects item generation | Instrument limitations | | C-013 | Involvement alone does not assign authority | Provisional | Participation varies by stage | Decision rights | | C-019 | Instrument validity has a scope | Provisional | Measured domains limit interpretation | Native limitations | | C-020 | Collaboration need not prove clinical benefit | Provisional | Process and outcome differ | Outcome measures | Drift affects evaluation, not what this fixed edition says. Software, current reporting standards, funder requirements, journals and terminology can change. No current software or journal recommendation is inferred from the 2017 discussion. Definitions and the described historical study procedures remain fixed. # Prompt 2 — Inspectional reading ## Book map The native structure has nine chapters and seven appendices. The publisher’s contents confirm the chapter titles. [Publisher contents](https://www.sagepub.com/shop/buy-a-book/designing-and-conducting-mixed-methods-research-3-241842). | Chapter | Purpose and contribution | Coverage | Hotspot | | --- | --- | --- | --- | | 1. The Nature of Mixed Methods Research | Defines the approach, rationales, skills and costs | Sampled U019–033 | Definition and conditions, C-001–005 | | 2. The Foundations of Mixed Methods Research | Locates history, philosophy and theory before technique | Sampled U038,043,044 | Worldview choices; not a complete philosophical audit | | 3. Core Mixed Methods Designs | Distinguishes fixed/emergent planning and three core interfaces | Sampled U050–059; U058 p/h/li blocks read throughout | Comparison, explanation and development, C-006–011 | | 4. Complex Applications of Core Mixed Methods Designs | Intersects core designs with experiments, cases, participation and evaluation | Sampled U070,074 | Complexity versus rigor and participant authority | | 5. Introducing a Mixed Methods Study | Aligns titles, aims and research questions | Sampled U088,091 | Integration question and sequencing | | 6. Collecting Data in Mixed Methods Research | Aligns sampling and data collection with design | Sampled U096,101,104,105 | Sample relationships and emergent follow-up | | 7. Analyzing and Interpreting Data in Mixed Methods Research | Separates strand analysis from integration and checks validity | Sampled U122–130 | Intent, procedure, display and inference | | 8. Writing and Evaluating Mixed Methods Research | Organizes reports and evaluates coherence | Sampled U143 | Actual mixing versus vocabulary | | 9. Advances in Mixed Methods Research | Describes developments at publication time | Sampled U148,153,154 | Displays and validity; software not evaluated | | Appendices A–G | Reproduces studies illustrating designs | A discussion sampled; B not independently read; C methods/limits sampled; D abstract/sampling sampled; E discussion opening; F methods/limits; G not read | Missing instrument domains and process/outcome limits | **Central project:** teach researchers how to design, conduct, integrate, report and evaluate a study with qualitative and quantitative components. The authors use a typology, methodological argument, published examples and writing scaffolds. The audience needs basic experience in both research traditions. The rhetoric teaches and advocates methodological coherence; it does not constitute a comparative trial of the textbook’s teaching effectiveness. **Provisional summary:** Mixed methods research in this edition involves two rigorous forms of inquiry, intentional integration, a design that organizes their relationship, and explicit theoretical and philosophical assumptions. The problem determines why you combine data. Convergent designs compare separate results; explanatory sequential designs use numerical findings to shape qualitative follow-up; exploratory sequential designs use qualitative inquiry to develop a feature that receives quantitative testing. Complex designs place these interfaces inside experiments, case studies, participatory work or evaluation. Sampling and validity must follow the design’s logic. A combined table can expose an inference, but the researcher still has to explain what the combination means. The sampled appendices show that local relevance does not assure complete instrument coverage and stakeholder collaboration does not prove clinical benefit. This is a practical reference whose application depends on skill, access, resources and an explicit account of what the second strand adds. Deeper reading should answer O01–04, O07–10, O13–15 and O17–19. Those twelve questions test the definition, value claim, selection decisions, authority and inferential limits. **Recommended strategy: selective deep read**, starting with your intended interface in chapter 3, then its sampling and analysis sections in chapters 6–7, and chapter 8’s evaluation guidance. Change to a full analytical read when a funded or participant-facing study needs a complete protocol. Stop adding strands when the second source has no unanswered question to address. # Prompt 3 — Classification and terminology Domain: social, behavioral and health research methodology. Genre: graduate methods textbook. Argument: a design-oriented synthesis with practical prescriptions, not a study establishing universal method superiority. Traditions named in sampled text include pragmatism, constructivist and postpositivist approaches, transformative work, evaluation and CBPR. Their names do not determine the quality of an actual study. All author definitions below are paraphrases; operational definitions and examples are system formulations. | Term | Author usage / source | Competing usage and risk | Operational definition; example / counterexample | | --- | --- | --- | --- | | Mixed methods | Two rigorous strands with integration, design and philosophy; U019 | Mere multiple techniques; label inflation | Show both analyses and their interface; survey plus explanatory interviews / interview coding alone | | Integration | Combine data/results for a stated intent; U122 | Any juxtaposition; no inference | Name procedure and resulting claim; merged comparison / separate reports | | Convergent | Merge separately analyzed strands; U057 | Qualitative triangulation; different traditions | Compare common constructs; parallel survey/interviews / numerical follow-up after instrument development | | Explanatory sequential | Quantitative then dependent qualitative work; U058 | A causal explanation; stronger inference than design permits | Trace result to interview sample/protocol; outlier follow-up / unrelated interviews | | Exploratory sequential | Qualitative, development, quantitative test; U059 | Two-stage shorthand that hides development | Trace themes to feature to test; new questionnaire / themes reported without testing | | Fixed design | Mixing planned at outset; U050 | Rigid implementation; no adaptation | Plan both strands; initial protocol / later addition after inadequacy | | Emergent design | Mixing added during study; U050 | Unreported opportunism | Document trigger and revision; new question / hidden hypothesis switch | | Core design | Three planning types; U055 | Complete taxonomy; forced classification | Use comparison/explanation/development intent / assume all projects use one pure form | | Complex design | Core interface within more components; U070 | Higher quality by complexity | Locate interfaces inside wider approach; CBPR plus convergent stage / more phases called better | | Connecting | Earlier results shape later phase; U058 | Shared participants alone; weak dependency | Trace sample and questions to result; outlier follow-up / reuse cohort with no analytic link | | Building | Qualitative findings shape quantitative feature; U059 | Any instrument drafting; no source trail | Record theme-to-item decisions; locally developed scale / off-the-shelf scale with no adaptation | | Joint display | Represent combined results; U122,153 | Proof by presentation; visual authority | Align construct, result, qualifier and inference; comparison table / decorative parallel columns | | Mixed methods question | Specifies integration; U091 | Two subquestions; missing combined task | Ask what comparison or follow-up establishes; explain unexpected scores / list two unrelated aims | | Validity | Threats to correct integrated inference; U129 | Quality, legitimation and inference quality; U128 | Test strand and interface threats; sampling audit / independent strand checks only | | Nested samples | One strand’s participants form subset of another; U101 | Independent validation; double-counting | Show subset relation; interviews within survey cohort / separate population called a subset | | CBPR | Community engagement across research stages; U074,244 | Participation as recruitment; erased authority | Record question-setting, feedback and action roles; shared interpretation / consultation without influence | Contested terms: mixed methods, validity and worldview vary across scholarly positions (U019,128,044). Completeness needs a specified question, and explanation needs a distinction between an account and identified causation. Participation varies from consultation to research-team membership (U074 e073). Category-error watchlist: correlation/causation is a live risk when interpreting follow-up accounts (C-010); description/prescription is a live risk in moving from a study example to universal guidance (C-003); individual/institutional levels require sample mapping (C-015); anecdote/population inference is a live risk in generalization from small local samples; intention/effect is a live risk for CBPR outcome claims (C-020); technical capability/practical result is a live risk for software/display claims. Absence of evidence/evidence of absence is present as a documented instrument-coverage risk in Appendix C. Identity/behavior, legal/moral and theological/historical categories are not applicable to the selected central argument. No accusation of those errors across unread text follows. # Prompt 4 — Claims inventory and dependencies EA means explicit author claim; Q3P means a claim by the reproduced article’s authors. Every row is a paraphrase. No RI statement is presented as an author sentence. Methodological and prescriptive rows use failure conditions for applicability; those conditions do not empirically falsify a definition. | Claim ID | Claim and source class | Type / scope | Testability and expected evidence | Falsifier or applicability defeater | Key term | Source / chapter | Linked question | | --- | --- | --- | --- | --- | --- | --- | --- | | C-001 | EA: Mixed methods requires rigorous qualitative and quantitative inquiry plus integration, design logic, theory and philosophy. | definitional; Authors’ working definition | Definition boundary | Find a study they explicitly accept despite no integration | mixed methods | U019 e036–039; S-01; 1 | O01 | | C-002 | EA: Use mixed methods when a research question needs something neither strand provides alone. | prescriptive; Study selection | Compare candidate designs | Show the second strand contributes no answer beyond a simpler design | rationale | U021 e004–006; S-02; 1 | O05 | | C-003 | EA: Combining data can produce a more complete understanding than either approach alone. | methodological; Conditional value of integration | Comparative study of inferences | Combined inference remains unchanged or worsens under mismatched constructs | completeness | U022 e004–005; 1 | O04 | | C-004 | EA: Researchers should gain competence in both traditions before conducting mixed methods. | prescriptive; Researcher preparation | Training guidance and project audits | Projects repeatedly fail despite claimed competence or succeed with supervised division of labor | competence | U031 e004–008; S-04; 1 | O12 | | C-005 | EA: Time, resources, personnel and coordination constrain feasibility. | practical; Project planning | Budgets and schedules | A proposed design cannot meet access, staffing or phase dependencies | feasibility | U032 e004–013; S-05; 1 | O12 | | C-006 | EA: Designs may be fixed in advance or emerge when one method becomes insufficient. | definitional; Design development | Explicit change record | A supposed emergent design has no traceable reason for adding a strand | emergence | U050 e004; S-06; 3 | O12 | | C-007 | EA: Three core designs organize comparison, explanation and development. | methodological; Authors’ typology | Match intent to procedures | A study’s intent and actual interface fail to fit its selected type | core design | U055 e004; U057 e004; U058 e004; U059 e004; 3 | O05 | | C-008 | EA: Convergent designs analyze two strands and merge results to compare or combine them. | definitional; Convergent design | Protocol and merged results | No comparison or combined inference appears after separate analyses | merging | U057 e029,043; 3 | O01 | | C-009 | EA: Divergent convergent-design results can generate insight and require investigation. | methodological; Discordant findings | Reanalysis and construct checks | Discordance comes entirely from a coding error or incompatible time windows | divergence | U057 e067; S-07; U238 e004; 3/Appendix E | O10 | | C-010 | EA: Explanatory sequential design uses quantitative results to plan qualitative follow-up and later integrated interpretation. | definitional; Explanatory sequence | Trace results to sampling and questions | Follow-up sample and protocol do not depend on initial results | connecting | U058 e031,045; S-08; 3 | O07 | | C-011 | EA: Exploratory sequential design uses qualitative findings to build a quantitative feature and test it. | definitional; Exploratory sequence | Item-development chain and testing | Items cannot be traced to exploration or the feature is never tested | building | U059 e004,038,043,074; S-09; 3 | O06 | | C-012 | EA: Complex applications contain more components and aren’t thereby more rigorous. | methodological; Experiments, cases, participation, evaluation | Design comparison | A complex label is used as evidence of quality without stronger procedures | complexity | U070 e005; S-10; 4 | O17 | | C-013 | EA: Participatory and social justice frameworks should shape the research process throughout. | prescriptive; Participatory applications | Decision rights and participant role records | Participants have no influence over questions, interpretation or action | participation | U074 e011,031,068,073; 4 | O13 | | C-014 | EA: Mixed methods questions should indicate the integration and align with the design. | prescriptive; Question writing | Question–method–inference correspondence | A stated integration question cannot be answered by the proposed interface | integration question | U091 e023,040,057–060; S-11; 5 | O11 | | C-015 | EA: Sampling relationships and phase dependencies affect the inferences available. | methodological; Data collection | Population and sampling map | Integrated conclusions generalize to people absent from both sampling frames | sampling | U101 e010,018,023; U104 e004; U105 e004; 6 | O09 | | C-016 | EA: Integration planning needs intent, procedures, representation and interpretation. | methodological; Integrated analysis | Four-part analytic plan | One of the four parts is absent and no integrated conclusion can be reconstructed | integration | U122 e005–013; S-12; 7 | O15 | | C-017 | EA: Validity includes both strands and threats specific to their integration and design. | methodological; Inference quality | Threat-to-inference audit | Both strands pass isolated checks but interface yields an incorrect inference | validity | U128 e011–013; U129 e004; U130 e004; 7 | O17 | | C-018 | EA: Mixed methods evaluation should check actual mixing and logical fit among the study’s elements. | prescriptive; Reviewing reports | Methods/results review | A study passes the proposed review without any evidence of mixing | logical fit | U143 e038–056; S-13; 8 | O01 | | C-019 | Q3P: The CVQ omits sexual violence, electronic media violence and stalking; its cross-context transfer needs validation. | descriptive; Enosh and colleagues’ reproduced instrument | Content and validation audit | Native items measure the omitted domains or external validation supports transfer | coverage | U197 e004; U199 e004; S-14; Appendix C | O19 | | C-020 | Q3P: The New Haven CBPR study documents collaboration but cannot establish clinical impact because it collects no transition outcomes. | descriptive; Greysen and colleagues’ study | Methods and limitations | The study includes measured clinical outcomes and compares interventions | outcomes | U244 e043,062,088; U246 e092,104; S-15–16; Appendix F | O14 | The textbook supplies procedural reasoning and selected examples for most guidance. It does not supply a comparative estimate of how often mixed methods outperforms one method across research settings. The reproduced articles are relevant demonstrations; selective quotation of them does not create independent replication. Definitions and prescriptions receive N/A in Prompt 7 where empirical grading would misstate their role. **Dependency graph:** C-001 is load-bearing: removing integration removes this book’s definition. C-002 supplies the rationale; C-007 selects the organizing intent; C-008/C-010/C-011 implement alternative interfaces; C-014/C-015 connect questions and samples; C-016/C-017 make inference inspectable; C-018 evaluates the whole design. C-003 is the weakest value bridge: fitting two strands does not guarantee better knowledge. C-005 can fail independently through resources; C-013 needs authority arrangements beyond technical integration. C-019/C-020 constrain transfer from particular examples and don’t overturn the core definitions. # Prompt 5 — Quantitative rigor policy This textbook contains empirical examples and discusses statistical work, but the review has no dataset that identifies a probability that its methodology improves research. No numerical posterior is invented. Definitions and prescriptions are N/A for probabilistic adjudication. Empirical claims use qualitative updates after Prompt 7, with the source and dependence listed there. Policy: begin with the provisional judgments in Prompt 1; examine the native procedure or reported limitation; update only the scoped proposition. Several chapters repeat one authorial method recommendation, so count them as one reasoning source. A reproduced article is a separate author team, but selected and interpreted by the textbook. Never treat numerical and qualitative reports from the same cohort as independent replications. Confidence vocabulary: unresolved when evidence does not choose among mechanisms; conditional when support depends on named design conditions; text-confirmed for directly inspectable statements about the supplied edition. These labels describe judgment, not measured probabilities. The template’s percentage bands are not populated because no defensible numeric likelihood is available. Calibration audit: false precision and correlated evidence are high risks if a posterior is assigned to C-003; selection bias is a high risk when exemplary published studies stand for all projects; tiny samples and unavailable denominators limit some qualitative transfer; publication and survivorship bias remain possible, unmeasured in this review. Base-rate neglect arises if successful examples replace the denominator of attempted projects. Proxy measures matter when collaboration substitutes for outcomes. P-values-as-effect-size and statistical-versus-practical significance are applicable cautions for Appendix B, whose full analysis is not read here. Motivated estimates are not measured; no reader motives are inferred. Stronger quantitative claims require the native tables, methods and data that this sample excludes. # Prompt 6 — Analytical scaffold and evidence ledger Use the chapter map in Prompt 2 as the analysis slots. Chapter 1 defines feasibility; chapter 2 places philosophy before method; chapter 3 chooses interfaces; chapter 4 places them in broader frameworks; chapter 5 names the combined question; chapter 6 maps samples; chapter 7 checks the resulting inference; chapter 8 checks reporting; chapter 9 records publication-era developments. Sources, claim ranges and coverage are attached to each slot. Reader response remains unavailable in every slot. Chapter 3’s sampled sections supply counterexamples through mismatch, divergence and incomplete advance specification (U057 e063–067; U058 e064–067; U059 e069–077). Chapter 4 recognizes different degrees of participation (U074 e073). Chapter 8 acknowledges competing quality orientations (U143 e059). Chapters 2 and 9 remain too sparsely sampled to assess counterexamples across each whole chapter. No section is labeled counterexample-free based on unread material. **Argument spine:** (1) research questions determine information needs; (2) each tradition offers distinct resources; (3) some questions require their relationship; (4) a design specifies that relationship; (5) integration must be carried through sampling, analysis, interpretation and reporting; (6) evaluate strand rigor and the validity of the interface. The weakest step connects distinct resources to a better answer (C-003). Repair it with a declared added inference, comparison against the single-strand inference and checks for shared bias. That narrows the benefit claim without discarding the framework. | Evidence ID | Claims | Type / source | Independence | Quality and limit | Native anchor | | --- | --- | --- | --- | --- | --- | | E01 | C-001–005 | Definitions and feasibility reasoning | One textbook author team | Direct for their position; no universal outcome test | U019,021,022,031,032 | | E02 | C-006–012 | Design procedures and challenges | Same team; not extra confirmation | Inspectable logic; effectiveness not quantified | U050,055,057–059,070 | | E03 | C-013–018 | Participation, questions, sampling, validity and evaluation | Same team synthesizing cited literature | Alternative positions reported; original studies not all checked | U074,091,101,122,128,129,143 | | E04 | C-009 | School research discussion | Appendix E’s Smith and colleagues, separate team | Opening reports intended and unexplained divergence; full methods unread | U238 e004 | | E05 | C-019 | CVQ limitations and item omission | Enosh, Tzafrir and Stolovy; separate team | Direct self-report about instrument scope; no replication | U197,199 | | E06 | C-020 | CBPR methods and limits | Greysen and colleagues; separate team | Collaboration documented; clinical outcomes absent | U244; U246 e092,104 | | E07 | C-010 | Depression-study limits | Wittink, Barg and Gallo; separate team | Patient perceptions and nonconcurrent assessments limit mechanism inference | U164 e004–005 | | E08 | C-013 | Gait-trial protocol sampling | Wiart, Rosychuk and Wright; separate team | Includes children/parents and decliners; protocol is not outcome evidence | U205 e004; U223 e004 | Five distinct reproduced author teams appear in sampled evidence E04–08. Their texts are separate from the textbook team; they are neither five replications of one benefit claim nor an external verification of their datasets. E05–07 tell against broad transportability and causal/outcome readings. Zero datasets are independently reanalyzed. Next reading order: chapter 3 interface → corresponding collection section → corresponding analysis section → Appendix example’s full methods/tables → chapter 8 criteria. Attention is high for measurement, sample relationships and combined inference; lower for publication-era resource lists. Before a stronger conclusion, obtain unread tables, inspect omitted cases and compare an alternative methodological account. Prompt 7 begins only after each scoped claim has a source, a relevant objection and an applicability boundary; all twenty rows below meet those conditions. # Prompt 7 — X-ray analysis and fair criticism **Strongest reconstruction:** the authors do not say you should always mix methods. They say to justify mixing for a specific information need, select a design, apply rigorous procedures to both strands, and make the interface and combined conclusion explicit. This reconstruction orders their teaching into a single chain; it is a system formulation, not a quotation. | Claim ID | Author position | Support summary | Grade | Strongest objection / counterexample | Current conclusion and repair | | --- | --- | --- | --- | --- | --- | | C-001 | Mixed methods requires rigorous qualitative and quantitative inquiry plus integration, design logic, theory and philosophy. | U019 e036–039; S-01; Native methodological formulation; empirical grade not assigned | N/A | Definitions vary across the field; example: A numerical coding exercise alone lacks a second strand | Use as this book’s definition; don’t claim universal consensus. | | C-002 | Use mixed methods when a research question needs something neither strand provides alone. | U021 e004–006; S-02; Native methodological formulation; empirical grade not assigned | N/A | Researchers can invent a rationale after collecting data; example: A descriptive frequency question needs no interviews by default | Require a named unresolved question before adding a strand. | | C-003 | Combining data can produce a more complete understanding than either approach alone. | U022 e004–005; Native procedures/limitations support the bounded proposition; no comparative replication | C | The text moves from complementary strengths to a broad benefit; example: Two biased samples agree while omitting the same group | Repair to potential gain when constructs, samples and integration fit; faithful but loses automatic superiority. | | C-004 | Researchers should gain competence in both traditions before conducting mixed methods. | U031 e004–008; S-04; Native methodological formulation; empirical grade not assigned | N/A | Mastery thresholds are unspecified; example: A team can distribute expertise without each person mastering both | Require competence across the team and accountable integration leadership. | | C-005 | Time, resources, personnel and coordination constrain feasibility. | U032 e004–013; S-05; Native procedures/limitations support the bounded proposition; no comparative replication | B | No quantified cost model accompanies the guidance; example: Two phases fit the calendar but participants cannot return | Adopt feasibility gates; no cost estimate follows from this book. | | C-006 | Designs may be fixed in advance or emerge when one method becomes insufficient. | U050 e004; S-06; Native methodological formulation; empirical grade not assigned | N/A | Emergence can conceal opportunistic analysis; example: Changing a hypothesis after seeing results without disclosure | Preserve decisions and revision dates; emergence does not excuse unexplained changes. | | C-007 | Three core designs organize comparison, explanation and development. | U055 e004; U057 e004; U058 e004; U059 e004; Native methodological formulation; empirical grade not assigned | N/A | A typology may force an adaptive project into a label; example: A project uses multiple cycles and needs a complex application | Use the three as planning scaffolds; not an exhaustive taxonomy. | | C-008 | Convergent designs analyze two strands and merge results to compare or combine them. | U057 e029,043; Native methodological formulation; empirical grade not assigned | N/A | Parallel timing alone does not establish integration; example: A report prints a survey beside an interview summary | Require an explicit comparison and inference from it. | | C-009 | Divergent convergent-design results can generate insight and require investigation. | U057 e067; S-07; U238 e004; Native procedures/limitations support the bounded proposition; no comparative replication | B | Novelty can be mistaken for truth; example: A scale measures trust while interviews ask about task completion | Investigate error and construct mismatch before treating discordance as discovery. | | C-010 | Explanatory sequential design uses quantitative results to plan qualitative follow-up and later integrated interpretation. | U058 e031,045; S-08; Native methodological formulation; empirical grade not assigned | N/A | Explanation can mean participant account or causal mechanism; example: Interviews suggest a reason but cannot exclude confounding | Describe explanatory accounts; causal identification requires its own design. | | C-011 | Exploratory sequential design uses qualitative findings to build a quantitative feature and test it. | U059 e004,038,043,074; S-09; Native methodological formulation; empirical grade not assigned | N/A | Selected qualitative themes may exclude unvoiced experiences; example: A rare harm never appears in the initial interviews | Audit the path from theme to item and record excluded dimensions. | | C-012 | Complex applications contain more components and aren’t thereby more rigorous. | U070 e005; S-10; Native methodological formulation; empirical grade not assigned | N/A | Added phases can conceal weak links; example: A seven-phase study has invalid measurement | Judge each strand and interface; complexity earns no quality credit. | | C-013 | Participatory and social justice frameworks should shape the research process throughout. | U074 e011,031,068,073; Native methodological formulation; empirical grade not assigned | N/A | Collaboration and justice are combined despite different commitments; example: Stakeholders advise on recruitment but cannot contest interpretation | Name who decides at each stage; participation is a continuum, not a label. | | C-014 | Mixed methods questions should indicate the integration and align with the design. | U091 e023,040,057–060; S-11; Native methodological formulation; empirical grade not assigned | N/A | A writing script can disguise a missing analytic procedure; example: A purpose statement says explain but there is no follow-up plan | Check the procedure that answers the question, not wording alone. | | C-015 | Sampling relationships and phase dependencies affect the inferences available. | U101 e010,018,023; U104 e004; U105 e004; Native methodological formulation; empirical grade not assigned | N/A | The same participant can contribute dependent evidence; example: Nested interview participants are treated as independent validation | Map identical, parallel, nested or multilevel samples and their limits. | | C-016 | Integration planning needs intent, procedures, representation and interpretation. | U122 e005–013; S-12; Native methodological formulation; empirical grade not assigned | N/A | A completed display can still support a poor inference; example: A visually tidy table compares unrelated constructs | Use four explicit checks; a display is a record, not proof. | | C-017 | Validity includes both strands and threats specific to their integration and design. | U128 e011–013; U129 e004; U130 e004; Native methodological formulation; empirical grade not assigned | N/A | The terminology remains contested; example: One sample excludes nonrespondents later described by another | Treat interface validity as a third responsibility; disclose competing vocabulary. | | C-018 | Mixed methods evaluation should check actual mixing and logical fit among the study’s elements. | U143 e038–056; S-13; Native methodological formulation; empirical grade not assigned | N/A | Mixed methods vocabulary may favor insiders; example: An unlabeled study integrates data but lacks field terminology | Weight observable procedures above labels; the authors acknowledge multiple evaluation orientations. | | C-019 | The CVQ omits sexual violence, electronic media violence and stalking; its cross-context transfer needs validation. | U197 e004; U199 e004; S-14; Native procedures/limitations support the bounded proposition; no comparative replication | B | The appendix defends Israel as a Western microcosm without testing transportability; example: A context where digital harassment dominates workplace violence | Accept documented omission; don’t infer omitted harms are absent. | | C-020 | The New Haven CBPR study documents collaboration but cannot establish clinical impact because it collects no transition outcomes. | U244 e043,062,088; U246 e092,104; S-15–16; Native procedures/limitations support the bounded proposition; no comparative replication | B | A plan and relationships are sometimes read as demonstrated health improvement; example: A successful task force forms while readmissions remain unmeasured | Accept collaboration evidence; clinical benefit stays untested in this study. | Distribution: A 0; B 4; C 1; D 0; F 0; N/A 15. N/A identifies definitions and recommendations whose role is methodological; it does not mean proof or rejection. B identifies bounded procedural or textual support. C-003 receives C because the broad added-understanding claim lacks a comparative performance test in the sampled material. No A grade is assigned because direct text checks establish what this edition says, not independent validation of a general research advantage. ## Qualitative updates | Claim | Prior | Evidence and dependence | Update / current judgment | | --- | --- | --- | --- | | C-001 | Boundary needed | Native definition, same author team | Text confirms the boundary; no empirical posterior | | C-003 | Added information need not improve inference | Rationale plus mismatches, all selected examples | No meaningful update to a universal benefit; condition on fit and added inference | | C-005 | Dependencies constrain feasibility | U032 resource and leadership checks | Moderately increases confidence in the planning caution; no duration/cost number | | C-007 | Typologies help and constrain | U051 and the authors’ changed names in U052 | Confirms that the taxonomy is a revisable scaffold | | C-009 | Discordance has several possible mechanisms | U057 challenges plus separate Appendix E discussion | Moderately increases confidence in investigating mismatch; no automatic insight claim | | C-010 | Accounts need causal limits | U058 plus separate Appendix A limitations | Supports a causal caution; participant explanations remain accounts | | C-011 | Local themes can miss dimensions | U059 plus C’s instrument limitation | Moderately increases confidence in auditing excluded domains | | C-013 | Involvement does not assign authority | U074 continuum and F’s strict-CBPR caveat | Confirms the need to inspect participant roles; no empowerment effect size | | C-019 | Validity has scope | Direct instrument limits and note | Strongly increases confidence in the omission finding for this instrument | | C-020 | Process need not prove outcomes | F explicitly collects no transition outcomes | Strongly increases confidence in keeping clinical impact unresolved | The dominant update inputs are explicit source limitations, not how often the textbook repeats a recommendation. No reader belief is changed on your behalf. Bias/fallacy scan: exemplary studies create a selection risk for C-003; this is an evidential limitation, not proof of deliberate cherry-picking. C-019’s Appendix C treats one national environment as a Western microcosm; that transportability argument lacks a cross-national test in the sampled text. C-018’s vocabulary check can create an insider-language bias if it outranks observable integration. A repair places procedures first. Broad completeness is repairable by narrowing the question, populations and added inference; the rhetorical cost is losing an automatic advantage claim. Calling the CBPR study proof of clinical benefit is not repairable as an outcome claim because outcomes were not collected; the faithful repair is to describe collaboration and planned change. Counter-observations in the authors’ favor: they allow single-method work (S-02), discuss time and skill constraints (S-04–05), concede convergence difficulties (S-07), distinguish complexity from rigor (S-10), report changing their own typology (U052), and accept diverse quality criteria (U143 e059). The appendices print limitations that weaken promotional readings. These observations rule out portraying the book as an unconditional endorsement of mixing. Attribution discipline: **Author says** the definition, design procedures and recommended evaluation checks. **Evidence shows** what the native text and reproduced methods disclose. **We infer** that an interface audit can help a documentation-research project. **We conclude** that the framework is useful when the combined question and sampling fit are explicit. **Still uncertain:** method superiority across projects, optimal training thresholds, transportability of examples and current software/reporting advice. # Prompt 8 — Systems and intellectual context The sampled historical account dates the consolidation of mixed methods literature to late-1980s/early-1990s sociology, evaluation, management, nursing, medicine and education (U038). That is the authors’ account of field formation, not proof that researchers never combined evidence earlier. U019 records competing definitions; U051 distinguishes typology-based and interactive approaches; U052 states that the authors changed design names and counts across editions. This third edition is one position in a developing field. Intellectual-lineage map: evaluation rationales and debates over philosophy → mixed methods definitions → design typologies → integration and validity frameworks. The textbook cites Greene, Tashakkori and Teddlie, Johnson and colleagues, Crotty, Mertens, Hesse-Biber and O’Cathain in the sampled discussion. Citation establishes engagement, not agreement with every proposition in those works. Their complete texts are not independently read here. Institutional settings shape what the book rewards: graduate proposals, research teams, journal review and funding applications (chapters 5–8). U029 e015 acknowledges publication outputs; U143 considers funders, teachers, researchers and participants as different evaluators. The economic constraint is staff, time, software and transcription cost (U032); no current prices are checked. The political and ethical question is who chooses the problem and interprets the answer, addressed in participatory work (U074). A documentation study combining usage logs and interviews is a **convergent system analogy**, not documented influence on this book and not a claimed native example. Contemporary automated coding or AI summaries are outside this edition’s verified advice. Current guidance would require fresh verification. Participant perspectives occur through reproduced studies; editorial selection and institutional research access shape the record. Voices outside the selected samples cannot be reconstructed from the textbook’s categories. # Prompt 9 — DEIB, multicultural and accessibility audit Structural inclusion appears where participants shape questions, revise instruments, interpret data and inform action (U074; U244 e052,062,088). Appendix F documents continuity of collaboration despite leadership changes (U246 e092). These are stronger mechanisms than a statement that researchers value inclusion. Their limits remain: the authors acknowledge that stricter CBPR could involve participants more, and the sample is 80% male (U246 e104). Representation does not establish shared control. | Transfer issue | Textual evidence | Limit / responsible adaptation | | --- | --- | --- | | Local instrument relevance | Exploratory sequence builds from participants’ experience, U059 | Map whose experiences inform items; validate again in a new setting | | Silence about harm | CVQ does not measure sexual violence, digital violence or stalking, U197/199 | Silence is a measurement gap; seek safe, consented input before adding domains | | Cross-national claims | CVQ developed in one national culture, U197 | Do not treat a microcosm analogy as tested transportability | | Housing and gender | F recruits one shelter/community, 80% male, U246 | Check women, families and people outside the shelter sampling frame before generalizing | | Children’s agency | Gait protocol includes both child and parent views and trial decliners, U223 | Preserve different goals; don’t replace a child’s perspective with a parent’s | | Research authority | Participation ranges from recruitment to team membership, U074 | Record who has decision rights and how disagreements survive reporting | The sampled social justice discussion names gender, racial/ethnic, class, disability and sexual-orientation perspectives (U074). The review cannot inventory demographic representation throughout unread chapters. Disability appears in a trial protocol; its outcome focus concerns walking, while the qualitative sample acknowledges that parents and children may hold different goals. No diagnosis, preference or mobility goal is assigned to an individual here. Accessibility: repeated procedures and flowcharts can help navigation, but the actual EPUB image figures and tables are not visually inspected. Screen-reader order, alt text, contrast and zoom remain unverified. This note uses headings, table labels and text descriptions beside diagrams; wide tables can still demand horizontal navigation. For study participation, U244 describes reading questions aloud and compensation. Those procedures address some literacy and time barriers; they do not establish language access, sensory access or freedom from institutional pressure for every participant. Check communication formats, consent and fatigue in a real protocol. Reading demand: syntax is mainly explanatory; reference density, methodological vocabulary, sampling logic and validity debates drive the difficulty. The system estimates graduate-level methodological prerequisites without a measured readability score. Supplementary perspectives should include full original studies, a participant-led account of authority and a rival account of mixed methods quality. These are categories for further reading, not invented bibliographic sources. # Prompt 10 — Embedded notes pipeline Fleeting questions: Is the second strand needed? Which initial result selects participants? What contradicts the combined story? What was excluded when items were drafted? Who can refuse or contest an interpretation? These are system prompts, not recorded reader reactions. ## Literature notes **LN-01 — The interface is the research task.** Creswell and Plano Clark make integration a core characteristic (C-001; S-01), then require intent, procedure, representation and interpretation (C-016; S-12). Their guidance moves from labeling a study to explaining the relation between its findings. Related permanent-note draft: [[#^pn-01]]. ^ln-01 **LN-02 — Sequence changes what you can plan.** Explanatory work chooses follow-up questions and participants after quantitative results (C-010; U058). Exploratory work must select qualitative findings, develop a quantitative feature and test it (C-011; U059). Both sequences need a record of deferred decisions and permissions. ^ln-02 **LN-03 — Relevance has a coverage boundary.** Appendix C’s authors acknowledge missing violence domains and a single-country instrument (C-019; S-14). An instrument can match expressed experiences and still omit unreported experiences. This inference concerns measurement coverage, not the frequency of those harms. Related permanent-note draft: [[#^pn-02]]. ^ln-03 **LN-04 — Process evidence stops before outcome evidence.** Appendix F records collaboration and community feedback (U244; S-15) and explicitly lacks measured transition outcomes (C-020; S-16). Collaboration is evidenced; clinical benefit is not evaluated there. Related permanent-note draft: [[#^pn-03]]. ^ln-04 ## Permanent-note drafts **PN-01 — An integrated result needs a traceable interface.** System synthesis: when two evidence streams answer one question, record which result changes the interpretation of the other. Confidence: conditional. Scope and conditions: applies to intentional comparison/explanation/development; does not require two methods for every inquiry. Support: [[#^ln-01]] and C-008/C-016. ^pn-01 **PN-02 — New measures inherit the blind spots of their development sample.** System synthesis: trace candidate dimensions to participants and seek omitted perspectives before claiming comprehensive coverage. Confidence: conditional, illustrated by one source instrument. Scope and conditions: instrument construction; not proof that all emergent measures fail. Support: [[#^ln-03]], C-011/C-019. ^pn-02 **PN-03 — Participation and outcomes need separate evidence.** System synthesis: document who influences decisions, then measure outcomes if you claim them. Confidence: conditional. Scope and conditions: research and evaluation projects; process improvements may matter even when later effects remain unknown. Support: [[#^ln-04]], C-013/C-020. ^pn-03 Structure note within this file: design choice → sampling relationship → integration → validity → reporting; PN-01 links the whole chain, PN-02 examines exploratory development, PN-03 examines participatory applications. Verified existing adjacency: [[4. Archives/Book reviews/The Craft of Research]] for question and argument construction; [[4. Archives/Book reviews/The Cambridge Handbook of Computing Education Research]] for study design; [[4. Archives/Book reviews/Research That Scales]] as a research-methods neighbor. These are navigational links only; their contents are not reread for comparison in this session. One-file authorization leaves reciprocal links to other notes unapplied. Attribution audit: sixteen quotations match native substrings; twenty claim paraphrases retain EA/Q3P labels; PN-01–03, the documentation analogy, the teaching case and proposed experiments are system creations. **Do not attribute to the authors:** reader motives, personal ratings, the three permanent-note drafts, any contemporary AI workflow, any claim of universal method superiority or proven clinical benefit from Appendix F. No RI claim is smuggled into the author inventory. # Prompt 11 — Teaching pack Audience: a researcher or documentation practitioner who knows basic survey and interview methods. Prerequisites: sampling frames, variables versus qualitative themes, and the difference between a participant account and a causal test. Objectives: choose one of the three core intents; identify the interface; state one combined inference; identify one threat or unmeasured outcome. Key vocabulary: convergent, explanatory sequential, exploratory sequential, connecting, building, joint display and validity. **Whole example — hypothetical, system-created:** a documentation team finds that a task-completion measure drops after a guide changes. It samples volunteers from the initial measurement cohort to ask what interrupted their tasks. Interviews identify two reported barriers: an unclear prerequisite and a navigation problem. The team maps each quantitative subgroup to its interview themes and writes what the follow-up helps explain. This is an explanatory sequence because the first result shapes selection and questions. It does not prove that the guide change caused the drop; timing, participant differences and other changes remain possible explanations. No user study is run here. Lesson sequence: inspect the complete example; locate the first-phase result and follow-up decision; contrast an exploratory plan that interviews people before drafting an instrument; create one joint-display row; state the inference and its limit. A learner chooses a case and does the work before receiving further assistance. No learner attempt or comprehension outcome is fabricated. Misconception checks: two data types without an interface do not satisfy this book’s working definition; interviews after a survey do not automatically establish causation; a complex design is not automatically stronger; a formatted display is not a validity result; participants on a recruitment list do not prove shared authority. Discussion questions: Which initial result merits follow-up? Who cannot return for interviews? Could both strands omit the same people? What would refute your interpretation? Who can contest the conclusion? Activity: write one integration question and a four-column row containing numerical result, qualitative finding, relationship and inference limit. Assessment: explain the sample relationship and one rival interpretation without using the design name as proof. Reflection: identify what you would leave unanswered when evidence is missing. Facilitator cautions: disclose the sampled coverage and unread tables; don’t teach C-003 as a guarantee or the hypothetical documentation case as source research. Discuss the violence-measurement limitation without soliciting personal disclosures. Avoid conflating housing status with identity or agency. Offer a written response, spoken response or diagram with a text equivalent; explain vocabulary at use, permit private reflection and provide a low-pressure stopping point. Translation or cultural adaptation should involve the affected participants’ terminology and priorities, not a literal transfer of a scale. Further reading begins with the complete appendices and chapter-specific original studies. No teaching outcome is claimed. # Prompt 12 — Actions, PARA, experiments and drift Actions to consider: write an integration-intent sentence; map both sampling frames; record which result selects follow-up; preserve discordance; audit an instrument’s omitted domains. Questions needing evidence: does the second strand change your inference, who controls decisions, and which outcomes are actually measured? Avoid adding a second method solely for a stronger label. **Most likely misuse:** treating the mixed methods label or a joint display as proof that an inference is better or that an intervention works. PARA routing: this completed review stays in Archives; source-bound methodology belongs conceptually in Resources; ongoing research-quality responsibilities belong in Areas if an existing responsibility uses them; nothing enters Projects because no reader project or commitment is supplied. All embedded drafts remain here under the one-file instruction. | Proposed experiment | Claim / hypothesis | Action and measure | Duration / stop | Review | | --- | --- | --- | --- | --- | | Interface audit on an existing public report | PN-01 / explicit interface makes its inference traceable | Compare its stated question, strands, interface and conclusion; count unsupported transitions | One reading session; stop if the report lacks enough methods to judge | At session end, before adopting the result | | Item-coverage rehearsal with invented data | PN-02 / theme-to-item tracing exposes excluded dimensions | Draft three items from fictional themes and list omitted domains; measure whether each item has a source and limit | One exercise; stop before collecting personal or sensitive data | After the draft, revise PN-02 if tracing adds no useful boundary | | Process/outcome claim audit | PN-03 / separate columns prevent overclaiming | Take one public case description and mark process evidence and outcome evidence; count claims exceeding observed measures | One session; stop when missing outcomes make assessment impossible | At session end | No experiment is executed, scheduled or assigned to you. Any future work involving people must use informed participation and ordinary disclosed conversation, with appropriate research governance for a formal study. Decision rules: adopt the audit when another reader can reconstruct the combined inference; adapt when the same rule needs a different design interface; test further when discordance has competing explanations; reject PN-01’s proposed audit if it adds paperwork without revealing any actual inferential dependency; defer outcome claims when outcomes are missing. Revise PN-02 if omitted dimensions arise from another mechanism than source selection; revise PN-03 if you fail to separate decision rights from reported participation. Drift monitoring, proposed: before using software, a funder template or publication advice, check its current official guidance; reassess terminology and standards when a new edition or disciplinary standard changes requirements. An annual review date of 2027-10-10 is a suggestion, not an automation. The historical text, native omissions and this edition’s definitions do not drift; their contemporary relevance can change. No live product recommendation is made. # Prompt 13 — Embedded visual thinking pack The diagrams encode system synthesis and uncertainty. A text-only comparison matrix and chapter map accompany them. Green marks the central proposition, blue marks concepts, orange marks cautions/actions, and purple marks connections. Color never carries meaning alone: node labels and dashed arrows retain the distinction. No person is depicted, so demographic representation cannot be inferred from these diagrams. Native EPUB figures remain uninspected; these maps do not certify their accessibility. ## Book and argument map Purpose: show where integration enters the book’s teaching chain. Nodes group definition, design, implementation and evaluation left to right. Sources: C-001–018; S-01,08–13. The dashed value link marks the conditional benefit claim C-003. ```mermaid flowchart LR A[Question and rationale: C-001–002] --> B[Choose comparison, explanation or development: C-007] B --> C[Map samples and interface: C-014–015] C --> D[Analyze, display and interpret: C-016] D --> E[Check strand and integration validity: C-017–018] D -. Conditional benefit, not guaranteed .-> F[Added understanding: C-003] classDef thesis fill:#bcb5e3,color:#241c4a,stroke:#126782 classDef concept fill:#8ecae6,color:#023047,stroke:#126782 classDef caution fill:#fee2c3,color:#241c4a,stroke:#fb8500 class A thesis class B,C,D,E concept class F caution ``` Text alternative: question and rationale → choose an integration intent → map samples and interface → analyze, display and interpret → check strand and integration validity. Added understanding remains a conditional benefit (C-003). ## Design and evidence decision map Purpose: locate the actual interface for each core design. Top-down branches represent alternatives, not a quality ranking. Sources: C-008–011; S-07–09. The warning asks the question that prevents a decorative display from becoming a conclusion. ```mermaid flowchart TD Q[What must mixing accomplish?] --> C[Compare findings] Q --> X[Explain initial results] Q --> Y[Develop a quantitative feature] C --> M[Convergent: analyze both, then merge] X --> N[Explanatory: quantitative results shape qualitative follow-up] Y --> O[Exploratory: qualitative themes shape development and testing] M --> I[State the combined inference and sample limits] N --> I O --> I I -. Check before claiming benefit .-> W[Did mixing answer a question one strand could not?] classDef concept fill:#8ecae6,color:#023047,stroke:#126782 classDef caution fill:#fee2c3,color:#241c4a,stroke:#fb8500 class Q,C,X,Y,M,N,O,I concept class W caution ``` Text alternative: compare findings through a convergent merge; explain initial results through connected qualitative follow-up; develop a quantitative feature from qualitative themes through an exploratory sequence. Each route must state its combined inference and sampling limits before claiming benefit. ## Stakeholder and limit map Purpose: separate participation mechanisms from outcome evidence in Appendix F. Group stakeholder roles around the interface; retain a dashed connection to unmeasured clinical benefit. Sources: C-013/C-020; S-15–16; U244 e043,088 and U246 e092,104. ```mermaid flowchart LR P[Shelter participants] --> Q[Question and survey feedback] R[Hospital and shelter partners] --> Q Q --> I[Shared interpretation and recommendations] I --> A[Collaboration and action planning documented] A -. No transition outcomes collected .-> U[Clinical benefit remains untested] classDef connection fill:#bcb5e3,color:#241c4a,stroke:#647baf classDef concept fill:#8ecae6,color:#023047,stroke:#126782 classDef caution fill:#fee2c3,color:#241c4a,stroke:#fb8500 class P,R connection class Q,I,A concept class U caution ``` Text alternative: shelter participants and hospital/shelter partners inform questions and survey feedback, then shared interpretation and recommendations. The source documents collaboration and action planning; it does not measure transition outcomes or establish clinical benefit. Comparison matrix: convergent = compare / merge / construct mismatch; explanatory sequential = explain / connect / follow-up selection; exploratory sequential = develop and test / build / omitted dimensions. This matrix is a terminology and action map as well as a design contrast. A separate causal-loop diagram is omitted because the source does not estimate feedback mechanisms; a timeline is omitted because sampled history does not establish a complete chronology; separate evidence, terminology and action diagrams would repeat these three maps. Excalidraw, slides and images are not created under the single-note instruction. ## Session completion check All Prompts 0–13 produce usable outputs. Adaptations: EPUB replaces unmatched Readwise material; all analysis is sampled; no numeric Bayesian posterior; literature/permanent-note drafts and maps stay embedded; no reciprocal hub edit; Amazon link remains null. Missing material ranked by consequence: unread native tables and full empirical methods; independent replication/comparative outcomes; reader highlights and purpose; live rendering/access checks. System assumptions concern reference-use purpose and the hypothetical teaching case, not personal motives or commitments. No source conclusion is drawn from unread sections. Rating and reading date remain null; session date records processing only. Next prompt is null because this chain is complete within the stated sample, not because every page has been read. Saved Markdown and source checks do not verify live Obsidian rendering or Sync.