# Lesson 2 — Research Questions, Theory, and the Literature (v3 expanded)

*Companion-podcast transcript • Sarah & Kiffer*  
*~5,400 words • ~30 min audio*

---

**Sarah:** Welcome back to Office Hours. I'm Sarah.

**Kiffer:** And I'm Kiffer. Today is Lesson 2 of the qualitative methods arc. Research questions, theory, and the literature. In Lesson 1 we set the foundation. What qualitative data analysis is, the three commitments, the four research goals, the five kinds of data. Today we move upstream of that, to the work that determines what every later step of your project can actually do for you.

**Sarah:** And the textbook is unusually emphatic about this point. The question is the single most consequential choice a researcher makes. Every later decision — sampling, instrument, analytic method, write-up — should be traceable back to a well-specified research question.

**Kiffer:** In the prior courses, you were typically given research questions to answer with established designs. Case-control, cohort, regression. In this course the question is yours to formulate. And the texture of your data — twenty interview transcripts collected with a semi-structured guide — both enables and constrains what you can defensibly ask.

**Sarah:** Okay, let's start with the first big distinction. Exploratory versus confirmatory.

**Kiffer:** Right. And the key thing here is that exploratory and confirmatory describe the epistemic position of the researcher at the start of the study, not the analytic technique. An exploratory question is one for which you don't yet have a candidate answer, or you have only weak ill-formed candidates. A confirmatory question is one for which you have a specific candidate answer — a hypothesis, a model, a theory-derived prediction — and you're testing it against data.

**Sarah:** Qualitative work has historically been most associated with exploration.

**Kiffer:** Because qualitative methods are particularly good at discovering categories. What counts as a thing, what its dimensions are, how it differs across people. Discovery is the natural mode when you don't yet know what to measure. But the textbook is careful not to collapse qualitative work into exploration. Grounded theory's later phases, analytic induction, qualitative comparative analysis — those are all confirmatory qualitative techniques. We'll come back to them.

**Sarah:** Quick callback. Last lesson you said the three commitments are systematic, transparent, replicable. Where does this exploratory-confirmatory declaration fit?

**Kiffer:** It's a transparency move. By declaring in advance what kind of question you're asking, you're letting the reader see one of the most consequential analytic decisions before the data come into the picture. It pre-commits you to a particular kind of write-up and a particular kind of standard. That's transparency in the methodological sense — the move is visible.

**Sarah:** Give listeners an example. Same dataset, exploratory version, confirmatory version.

**Kiffer:** Sure. Exploratory: "What dimensions do British Columbian adults use when they describe their own experiences of loneliness?" Confirmatory: "Cacioppo and colleagues propose that loneliness is qualitatively different from being alone. Do the twenty transcripts in our dataset support, complicate, or contradict that distinction?" Same twenty transcripts. The first asks the data to teach you the categories. The second asks the data to evaluate a category already proposed in the literature.

**Sarah:** And you owe the reader an explicit statement of which posture you're taking.

**Kiffer:** That's the methodological discipline. You declare your posture. You don't drift between them silently.

**Sarah:** And the practical implication of declaring posture is what.

**Kiffer:** Two things. It shapes how your reader interprets your findings. A reader who knows you're doing exploratory work will read a finding as a candidate hypothesis. A reader who knows you're doing confirmatory work will read the same finding as evidence for or against a specific prior claim. Same finding, different epistemic weight. The posture declaration tells the reader what kind of weight to give your conclusions. The second implication is internal. It disciplines your write-up. Confirmatory studies have to defend the prior. Exploratory studies have to defend the absence of one. You can't slide between them.

**Sarah:** And you sometimes see papers that slide. Where the introduction reads exploratory but the findings section delivers a confirmatory verdict.

**Kiffer:** That's a methodological failure, and reviewers catch it. The textbook treats it as a posture-confusion problem. The cure is to declare upfront and to stick.

**Sarah:** Okay, the second tool in this section is the four screening questions. These are the textbook's filter for whether a question is ready to be a research question.

**Kiffer:** Right. Before you commit, you run the candidate question through four screens. Each one is a different kind of feasibility check. A question that fails any of the four is not yet a good question.

**Sarah:** Walk me through them.

**Kiffer:** Personal interest first. Will you still care about this question in month eleven of a twelve-month project? The failure mode here is picking a topic because it seemed publishable, not because you wanted to know the answer. Qualitative research is long, slow, recursive work. You're going to live with this question. If you don't actually care, you'll cut corners.

**Sarah:** Second screen.

**Kiffer:** Empirical answerability. Can the question, as stated, be answered with data — any data? The failure mode here is questions about value rather than fact. "Is loneliness a moral failing?" is not an empirically answerable question. "How do participants morally frame their own loneliness?" is.

**Sarah:** Third.

**Kiffer:** Resources. Do you have the time, money, access, language, and skills the question requires? The failure mode is the question that would need two hundred ethnographic interviews with policymakers in three provinces. Beautiful question. Not a year-long graduate capstone.

**Sarah:** And fourth.

**Kiffer:** Ethics. Can the study be conducted in a way that respects participants and meets the standards of the Tri-Council Policy Statement, TCPS two? The failure mode is the question that requires you to elicit traumatic disclosures without offering support, or that puts participants at risk of identification.

**Sarah:** Notice what's not on the list. Publishability.

**Kiffer:** Right. The textbook is critical of using publishability as a screening criterion, because it tends to push researchers toward fashionable questions rather than important ones. Many genuinely valuable questions are not currently fashionable. Many fashionable questions are intellectually thin. The four screens are a filter for whether the question can be done well, not for whether it will be celebrated.

**Sarah:** I want to add a small thing about the empirical answerability screen. Because it sounds dry but it's where a lot of questions die that students don't realize have died.

**Kiffer:** Yeah. The screen has two parts that students sometimes conflate. First, is the question, in principle, answerable with any data anywhere? Second, is it answerable with the data you have? The first kills value questions and pure-definition questions. The second kills questions that require data you can't access. A question can pass the first and fail the second, and the failure mode is sneaky because the question feels rigorous — it sounds like a real empirical question, it just isn't tractable with the corpus in front of you.

**Sarah:** Example?

**Kiffer:** "How does loneliness change over the course of a year of bereavement?" That's an empirically answerable question in principle. But twenty cross-sectional interviews collected at one time point can't answer it. You'd need either longitudinal data or retrospective accounts that explicitly traced the change. The capstone dataset doesn't give you that. The question passes the first part of the answerability screen and fails the second.

**Sarah:** And the fix is to narrow.

**Kiffer:** Right. The narrowed version might be "how do bereaved participants in our dataset describe the temporal shape of their loneliness — onset, intensity, change over time — as they recount it retrospectively?" That's answerable with the data. It's a different question. But it's a real one.

**Sarah:** Many students arrive at this course with a topic, not a research question. The four screens are most useful as a movement from topic to question.

**Kiffer:** Exactly. "Loneliness" is a topic. "How do recently-bereaved older women in British Columbia describe the shape of their loneliness, and what social, material, and embodied features of their daily lives do they identify as constitutive of it?" is a research question. The second one specifies who, what, where, and what kind of answer it expects.

**Sarah:** Let me push on the personal interest screen, because I think it's underestimated.

**Kiffer:** It is. Personal interest sounds like the softest of the four, the one a serious researcher might want to override. The textbook's argument is the opposite. Personal interest is what gets you through the long middle stretch of a qualitative project, the months where the coding is tedious and the patterns are not yet visible. If you don't actually care about the answer, you'll start defaulting to the patterns that fit what you wanted to find. Personal interest is a discipline tool. It keeps you honest because you have an investment in actually knowing.

**Sarah:** And it works the other way too. If you care too much, in a way that's pre-committed to a particular answer, that's also a problem.

**Kiffer:** Yes. The screen is personal interest, not personal verdict. You have to want to know, not want to prove. The positionality memo from last week is partly a check on this. When you write out your priors before you read the data, you can later notice if your findings are suspiciously congruent with them.

**Sarah:** And before we move on, a word about what the dataset constrains.

**Kiffer:** Worth saying explicitly. Twenty transcripts means questions about prevalence are off-limits. You cannot ask "how common is loneliness in BC adults?" — there's no answer the data can give you. Questions that require dense subgroup contrasts are also out. A question that splits the dataset fifteen ways will have one or two transcripts per cell, which is not enough to characterize anything. Questions about kinds, processes, and meanings are well-supported. What kinds of loneliness exist, how is each kind structured, how do people come to recognize their loneliness, what does loneliness signify.

**Sarah:** And there's a second dataset constraint, which is the interview guide.

**Kiffer:** Right. Topics the guide asked about — definitions of loneliness, triggers, coping, social world, stigma, policy preferences — will be richly represented across transcripts. Topics the guide did not directly ask about — clinical symptomatology, family-of-origin attachment styles, neurobiological correlates — will appear only sporadically. A defensible question concentrates on what the guide elicited.

**Sarah:** Okay, second big section of the lesson is theory and paradigms. Walk me into this carefully because it can feel abstract.

**Kiffer:** It can. And I'll say upfront that the paradigm conversation may feel uncomfortable if your training has been quantitative. Quantitative epidemiology rarely makes you say out loud that you believe in an external reality knowable through measurement, because almost everyone in the room is already a postpositivist. Qualitative research draws practitioners from very different positions, so the convention is that you declare yours.

**Sarah:** Start with what a paradigm is.

**Kiffer:** A paradigm is a bundle of three commitments. Ontology, which is "what kind of thing is the social world?" Epistemology, which is "how can we know about it?" And axiology, which is "what role do values play in research?" The bundles aren't all internally consistent in every detail, but they cohere enough that researchers within a paradigm tend to agree on what counts as a good study.

**Sarah:** And the textbook organizes the contemporary landscape into four paradigms.

**Kiffer:** Postpositivism, interpretivism, critical paradigms, and pragmatism. Walk through each briefly.

**Sarah:** Postpositivism first.

**Kiffer:** The view that an external reality exists and is, in principle, knowable. Strict positivism — which is rare today — treats social facts as fully observable. Postpositivism accepts that our access to reality is fallible and shaped by our instruments, while still treating reality as the target. A postpositivist qualitative study of loneliness would treat participants' accounts as imperfect windows onto an underlying phenomenon, and would prioritize procedures — intercoder reliability, triangulation, transparency — that increase the credibility of inference from accounts to phenomenon.

**Sarah:** Interpretivism, sometimes called constructivism.

**Kiffer:** The view that the social world is constituted through meaning-making, and that the researcher's job is to understand those meanings rather than to extract underlying facts behind them. An interpretivist study of loneliness would treat the accounts as the data — the loneliness of interest is the loneliness as constructed in talk, not a hypothetical thing behind the talk. Charmaz's constructivist grounded theory is the most widely cited contemporary interpretivist methodology in health research.

**Sarah:** Critical paradigms next. This is where feminist, anti-racist, decolonial, Marxist, critical-realist all sit.

**Kiffer:** Right. They share the commitment that research is not neutral and that part of the researcher's job is to make visible the structural conditions that produce the phenomenon. A critical loneliness study might foreground how policy choices — housing affordability, immigration restrictions, healthcare under-funding, long-term-care austerity — produce specific patterns of isolation, and would treat participants' accounts as evidence about structural conditions, not only about individual experience.

**Sarah:** And the fourth.

**Kiffer:** Pragmatism. A meta-stance that picks methods to fit the question rather than committing in advance to one paradigm. Pragmatist researchers mix methods, draw on different paradigms in different parts of a project, and ground decisions in "what works for this question." Pragmatism is common in applied health-services and implementation-science research. The critics worry it can become an excuse for non-disclosure. The defenders argue that strict paradigm-fidelity can produce studies that are philosophically clean but practically irrelevant.

**Sarah:** Same loneliness question, four different framings.

**Kiffer:** Yeah, and it's worth running through. A postpositivist framing would be "what patterns in loneliness experience can be inferred from these twenty transcripts, with what credibility, accounting for the fallibility of self-report?" An interpretivist framing would be "how do these participants construct loneliness in their own narratives, and what meaning-making work are they doing in the telling?" A critical framing would be "what structural conditions — immigration policy, housing precarity, racialization, gendered caregiving — are visible in these accounts, and how do they pattern who experiences which kinds of loneliness?" And a pragmatist framing would be "what can these transcripts contribute to the design of a loneliness intervention or the refinement of an existing measurement instrument, drawing on whichever analytic moves are most fit for purpose?"

**Sarah:** All four are defensible. The point is that they answer different questions and require different analytic moves.

**Kiffer:** That's the move. And honestly, for most students, the paradigm conversation is less about choosing the correct one and more about recognizing the one you're already in and being able to defend it. Most epidemiology-trained students arrive postpositivist by default. That is a defensible position. But it is a choice, not a neutral starting point, and your methods section should acknowledge it.

**Sarah:** What's the cost of just staying in your default postpositivism?

**Kiffer:** A postpositivist reading tends to flatten the meaning-making work participants do in interviews — the metaphors, the silences, the refusals — into noise around a signal, when those features may themselves be the most informative data. A line like Amira saying she experiences "wahda" — an Arabic word with no clean English equivalent — gets read as a translation problem in postpositivism. In interpretivism, the word itself is the finding.

**Sarah:** That's a vivid example. Okay, let's move to the literature. The third big section.

**Kiffer:** In the prior courses you learned to conduct a systematic review for quantitative evidence. PRISMA, PICO, MeSH terms, risk-of-bias instruments. The literature search for qualitative evidence works on a similar logic but with important differences. The searchable concepts are different, the databases are partly different, the inclusion criteria are different, and the synthesis methods are very different.

**Sarah:** Start with what's the same.

**Kiffer:** The basic discipline of a systematic search transfers directly. You specify your search question in advance using a structured frame. You search multiple databases. You document your search strings, dates, and result counts. You apply inclusion and exclusion criteria with an audit trail. You screen titles and abstracts, then full text, with a second reviewer if possible. You extract data into a structured table. You appraise the quality of included studies. The PRISMA flow diagram — or its qualitative variants like eMERGe for meta-ethnography and ENTREQ for transparency in synthesis reporting — is still the convention.

**Sarah:** And what's different.

**Kiffer:** Four things. First, the search terms. Qualitative studies are not always indexed with reliable terms like "qualitative" or "interview" in their titles or abstracts. The Cochrane Qualitative and Implementation Methods Group has published validated search filters for several databases. Use them rather than improvising.

**Sarah:** Worth naming SPIDER explicitly. Because PICO — population, intervention, comparison, outcome — is the structure students arrive with.

**Kiffer:** Right. SPIDER stands for Sample, Phenomenon of Interest, Design, Evaluation, Research type. Sample replaces Population. Phenomenon of Interest replaces Intervention. The shift signals that qualitative work is often not studying an intervention applied to a population. It's studying an experience or a phenomenon present in a sample. Design and Research type, in SPIDER, are about whether you're including interviews versus focus groups versus ethnography. PICO doesn't have a slot for that. SPIDER was developed by Cooke, Smith, and Booth in twenty twelve specifically because qualitative search questions kept getting forced into PICO and losing precision in the process.

**Sarah:** And one practical tip on the search side.

**Kiffer:** Always use a validated qualitative search filter if your database has one. MEDLINE has several. Improvising your own filter will usually produce both false positives — papers that mention "qualitative" in passing — and false negatives — papers that don't tag themselves as qualitative even though they are. The validated filters have been benchmarked against gold-standard reference sets. Use them.

**Sarah:** Second.

**Kiffer:** The databases. MEDLINE and Embase still matter, but CINAHL, the nursing database, plus PsycINFO and Sociological Abstracts and ASSIA — the Applied Social Sciences Index and Abstracts — are particularly important for qualitative health research. And grey literature, meaning theses, reports, working papers, is more important in qualitative reviews because qualitative studies are over-represented in unpublished form.

**Sarah:** Quick concrete on the SPIDER tool. Let's run loneliness through it.

**Kiffer:** Sure. Sample equals adults. Phenomenon of Interest equals experiences of loneliness. Design equals semi-structured interviews or focus groups. Evaluation equals themes, meanings, descriptions, conceptual frameworks. Research type equals qualitative. Take that to a database, combine it with a validated filter, and you have a defensible starting search.

**Sarah:** And one of the things students often miss is the importance of grey literature for qualitative work.

**Kiffer:** Yeah. Many of the best qualitative health studies are theses or organizational reports rather than indexed journal articles. ProQuest Dissertations and Theses Global is worth searching. So are the publication outputs of major qualitative health-research centres — the National Centre for Research Methods in the UK, the various Canadian qualitative-health collaboratives. If you only search MEDLINE you'll miss real work.

**Sarah:** Third.

**Kiffer:** The quality appraisal tool. The Cochrane risk-of-bias tool is not the right instrument for a qualitative study. The Critical Appraisal Skills Programme — CASP — qualitative checklist, and the Joanna Briggs Institute qualitative checklist, are the most widely used purpose-built tools.

**Sarah:** And fourth.

**Kiffer:** The synthesis. You don't pool effect sizes. Instead there's a whole literature on qualitative-evidence synthesis methods, and they differ in epistemology and in how aggressively they reduce findings.

**Sarah:** Walk listeners through the main ones briefly.

**Kiffer:** Five approaches dominate. Meta-ethnography, from Noblit and Hare in nineteen eighty-eight, translates concepts across studies to produce a third-order interpretation. It's the most interpretive end. Thematic synthesis, from Thomas and Harden, codes the findings of included studies, develops descriptive themes, then generates analytical themes that go beyond any single study. Qualitative meta-summary, from Sandelowski and Barroso, aggregates findings and computes effect-size analogues — frequency and intensity — across studies. That's the most quantitative end. Framework synthesis uses an a priori conceptual framework to organize findings, useful when there's an existing theory the synthesis aims to refine. And critical interpretive synthesis, from Dixon-Woods and colleagues, mixes systematic searching with interpretive reading. Designed for complex contested topics where a tidy synthesis would be misleading.

**Sarah:** That's a real spread.

**Kiffer:** It is. And the methodological choice — which synthesis approach you use — is downstream of your research question and your paradigm. A postpositivist reviewer might lean toward thematic synthesis or qualitative meta-summary. An interpretivist reviewer might lean toward meta-ethnography. A critical reviewer might lean toward critical interpretive synthesis. The framing of the synthesis enacts the paradigm.

**Sarah:** Now, for the capstone in this course, the deliverable is not a full synthesis.

**Kiffer:** Right. The capstone literature framing is more modest. Eight to twelve cited sources. Summarize what is known from qualitative research about your specific corner of the loneliness literature. Identify the gap your study would address. Locate your candidate research question in that landscape. One page, single-spaced, with citations.

**Sarah:** And the gap statement is where students struggle.

**Kiffer:** Yes. The strongest gap statements name a specific population, a specific phenomenon, or a specific comparison that the existing literature has under-described and that your dataset is well-positioned to address. The weak version is "more research is needed on loneliness." The strong version is "the qualitative loneliness literature is rich on older bereaved adults and is increasingly attentive to immigrant loneliness, but it has not yet adequately characterized the loneliness inside companionship — the experience of being lonely with a partner present, an adult disabled child present, or a dementia-affected parent present. The dataset contains three transcripts that describe this configuration directly. A focused analysis could contribute a defensible typology of companioned loneliness to the literature." That's a gap statement.

**Sarah:** Specific, dataset-grounded, and pointed toward a contribution.

**Kiffer:** Right. And honestly, the gap statement is one of the most important sentences in your eventual paper. Reviewers read it carefully.

**Sarah:** Let's also briefly cover ethics, because that's threaded through the lesson.

**Kiffer:** Quick walk-through. Even with the synthetic dataset, your methods section is expected to address what the ethical considerations would be for a real-world version of your study, under the TCPS two framework. Five areas. Free and informed consent — what would participants be told, how would consent be obtained for sensitive topics, how would withdrawal work? Confidentiality — how would identifying details be removed from transcripts, what's your data-storage plan, how will quoted excerpts be presented? Risk of harm — what's the risk of re-traumatization from an interview about loneliness, what support resources would be provided? Justice in subject selection — who is included, who is excluded, on what grounds? And researcher obligations of care — what does the researcher owe a participant who discloses something distressing during an interview?

**Sarah:** The synthetic dataset doesn't trigger an actual ethics review, but the methods section is the place where you demonstrate that you know what one would require.

**Kiffer:** Exactly. It's a piece of the transparency commitment. You're showing the reader that you understand the ethical infrastructure your design would sit inside if it were a real study.

**Sarah:** And one more thing on ethics before we move on. The synthetic dataset removes some real-world risk, but it doesn't remove the obligation to think through what a real version would look like.

**Kiffer:** Right. And there are pedagogical reasons for that obligation. Most students in this course will, at some point in their careers, run a real qualitative study. The methods-section discipline of working through TCPS two considerations is portable. The students who get it right on the synthetic dataset are practiced when they meet a real REB application. The students who skip it on the synthetic dataset hit the real REB cold and produce thin applications.

**Sarah:** And REBs in Canada are increasingly attentive to qualitative-specific concerns.

**Kiffer:** They are. Twenty years ago, a Canadian REB might have applied a basically biomedical framework to a qualitative interview study and missed most of what matters. Today, most institutional REBs have members who understand qualitative-specific concerns — the open-endedness of consent in emergent topics, the risk of identification from rich descriptive detail, the obligations that follow disclosure. The TCPS two revisions have helped. The framework has caught up.

**Sarah:** Last section of the lesson is the worked example. Topic to capstone question.

**Kiffer:** Five steps. Let me walk through it. A typical student arrives with a topic like "loneliness in older people." Step one is from topic to question stub. What about loneliness in older people interests you? Becomes "how older people describe the experience of loneliness." That's closer to a question, but still vague — "describe" could mean anything from clinical phenomenology to social-network mapping.

**Sarah:** Step two.

**Kiffer:** Specify the kind of answer. What do you want to know that you don't now know? Becomes "what dimensions and triggers of loneliness do older participants name in their own accounts, and how do these compare to younger participants' accounts?" Now the question has a comparison built in and is asking for content the dataset can provide.

**Sarah:** Step three.

**Kiffer:** Ground the question in the dataset. Can the twenty transcripts actually answer this? Five of the twenty are over sixty-five. Five is enough for descriptive characterization but not for fine-grained subgroup work. The question is dataset-feasible if you accept that "older" will be defined by those five and that comparison to younger participants will be at the level of patterns, not statistical contrasts.

**Sarah:** Step four.

**Kiffer:** Locate in the literature. What can your capstone add? In this case, a defensible answer is that the qualitative loneliness literature has studied life-stages largely separately. A single methodologically consistent dataset that lets you compare across life-stage is a contribution.

**Sarah:** And step five.

**Kiffer:** Run the four screens. Personal interest — you do care. Empirical answerability — the dataset can support the question with stated caveats. Resources — twelve weeks is sufficient. Ethics — the dataset is synthetic, so the question is REB-clean for instructional purposes. Final version: "How do older British Columbian adults — age sixty-five and over — describe the dimensions and triggers of their loneliness, and in what ways do their accounts pattern differently from those of younger adults in the same dataset?"

**Sarah:** That's a question another reader could predict what the methods section will say from.

**Kiffer:** Right. Specific population. Specific kind of answer expected. Comparison structure built in. Dataset-feasible. That's the target.

**Sarah:** Let me try to synthesize the lesson into takeaways. First, exploratory and confirmatory describe the researcher's epistemic position, not the analytic technique. Qualitative work can do both, and you owe the reader an explicit declaration of which posture you're taking.

**Kiffer:** Second, the four screens — personal interest, empirical answerability, resources, ethics — are the filter for whether a candidate question is ready to be a research question. Publishability is not on the list. The textbook is critical of letting it sneak in as a fifth screen.

**Sarah:** Third, a paradigm is a bundle of commitments about ontology, epistemology, and axiology. The four contemporary paradigms — postpositivism, interpretivism, critical, pragmatism — each shape what the same dataset can be made to say. Most students arrive postpositivist by default and need to acknowledge that as a choice.

**Kiffer:** Fourth, qualitative-evidence literature search uses the same discipline as quantitative review but different tools. SPIDER instead of PICO. CASP instead of Cochrane risk-of-bias. CINAHL and PsycINFO in addition to MEDLINE. And the synthesis methods — meta-ethnography, thematic synthesis, qualitative meta-summary, framework synthesis, critical interpretive synthesis — are profoundly different from meta-analysis.

**Sarah:** Fifth, a defensible research question is specific, dataset-feasible, and forward-clear. A reader should be able to predict in broad terms what your methods section will say from the question alone.

**Kiffer:** Sixth, TCPS two ethics applies to your methods section even with a synthetic dataset. Consent, confidentiality, risk of harm, justice, and obligations of care are still the five areas you address.

**Sarah:** Seventh, the upstream design work feels disproportionate to its concrete output — a paragraph and a page — but it determines what every later step of the course can do for you. A vague question produces a vague methods section, a vague analysis, and a vague paper. A precise question makes everything that follows possible.

**Kiffer:** And eighth, a callback to last lesson. The three commitments — systematic, transparent, replicable — are already showing up in this work. Your question, your paradigm, your literature framing, and your ethics treatment are all pieces of the transparency commitment. You're declaring in advance what your study is going to do and what stance it's working from.

**Sarah:** Anything you want listeners to carry into the next lesson?

**Kiffer:** One thing. The work this week feels like writing. Like just a paragraph and a page. But the paragraph and the page are infrastructure. The rest of the term will lean on them constantly. The students who do this week well, who really specify their question and name their paradigm honestly, have a much easier time in the analysis weeks.

**Sarah:** Next lesson is sampling. The textbook's argument that probability and nonprobability samples were built to answer different questions, and that the saturation logic of qualitative sample size is not a degraded version of the power calculation — it's a different question entirely.

**Kiffer:** Before then, draft your candidate research question, run it through the four screens, do a focused literature search, write the one-page framing, and name your paradigm.

**Sarah:** Thanks for listening. We'll see you in Lesson 3.

**Kiffer:** Take care of yourselves. See you in class.