Research path
Research methods and scholar examples Research training beyond introductory study: evidence synthesis, causal design, modelling, qualitative analysis and reproducible experiments.
Start with your research question, choose a method, study original materials and produce work others can examine. Sources are mainly in English, with bilingual reading tasks.
Six methods workbooks Scholars, courses and sources YouTube and talks
Choose a research task The workbooks and review criteria are platform-designed exercises for projects or group discussion, not assignments set by the source authors.
Questions and contributions: distinguish competing explanations Compare candidate questions, map claims to evidence and draft a proposal that can be challenged.
Method · example · deliverable · review criteria → Evidence synthesis: protocols, bias and disagreement Move from rerunnable searches to synthesis; separate reporting, study quality and comparability.
Method · example · deliverable · review criteria → Causal design: estimands, DAGs and sample justification Define the target trial and identification assumptions before estimation, then simulate where the design fails.
Method · example · deliverable · review criteria → Bayesian workflow: generative models, diagnostics and revision Connect priors, sampling diagnostics, predictive checks and sensitivity analysis in an auditable workflow.
Method · example · deliverable · review criteria → Reflexive thematic analysis: from extracts to interpretation State epistemology and positionality, trace codes into themes and revise interpretations using counterexamples.
Method · example · deliverable · review criteria → Reproducible experiments: baselines, fair comparisons and failure audits Fix splits, comparison budgets and sources of variability so another researcher can rebuild the main result.
Method · example · deliverable · review criteria → Scholar examples and professional resources Prerequisite level All Advanced · subject foundations Graduate · methods prerequisites
11 resources found
Questions and arguments · Papers and scholar essays · Advanced · subject foundations
Uri Alon · 2009
Reading target A researcher's perspective on problem choice and the experience of doing science.
Prerequisites A candidate topic and familiarity with representative papers.
Platform-designed exercise Compare three questions by existing answers, uncertainty, discriminating evidence and resource bottlenecks.
Method boundary An experience-based essay, not a validated predictor of publication success.
Access Lab reading list is public; the paper may require institutional access. Open workbook and review criteria →
Source checked 2026-10-25 Statistical modelling · Video and course · Graduate · methods prerequisites
Richard McElreath · 2023
Reading target Focus on lecture 6, Good & Bad Controls, and lecture 12, Multilevel Models; the author links the full course.
Prerequisites Probability, regression, R and willingness to simulate data.
Platform-designed exercise Simulate grouped data; compare complete, separate and partial pooling, including uncertainty for small groups.
Method boundary Adjustment depends on causal structure; complexity or better fit does not establish causality.
Access Public lectures, slides and code; textbook obtained separately. This selection uses the fixed 2023 edition. Open workbook and review criteria →
Source checked 2026-10-25 · Video addresses verified from author pages; recordings were not reviewed end to end. Statistical modelling · Papers and scholar essays · Graduate · methods prerequisites
Andrew Gelman, Aki Vehtari et al. · 2020
Reading target Model building, checking, computational diagnostics and comparison as an iterative workflow.
Prerequisites Bayesian inference, MCMC basics and a probabilistic programming tool.
Platform-designed exercise Submit prior and posterior predictive plots and a revision log tying each change to a failure.
Method boundary Passing diagnostics supports computation and checking, not the sampling or identification assumptions.
Access Open preprint; use after a probability course. Open workbook and review criteria →
Source checked 2026-10-25 Causal identification · Textbook · Graduate · methods prerequisites
Miguel A. Hernán & James M. Robins · 2020 · 作者页持续修订 / living author edition
Reading target Progresses from causal questions without models to modelling and complex longitudinal data.
Prerequisites Conditional probability, regression and observational study basics.
Platform-designed exercise Write a target-trial table: population, intervention, comparator, time zero, outcome and estimand.
Method boundary Observational data require explicit identification assumptions; more covariates do not guarantee less bias.
Access Book, data and associated code provided free by the author. Open workbook and review criteria →
Source checked 2026-10-25 Evidence synthesis · Handbook and guideline · Graduate · methods prerequisites
Cochrane methods authors · 6.5 · 2024
Reading target Evaluate bias domains for a specific randomized-trial result and document the reasons.
Prerequisites Study designs, effect measures and systematic review concepts.
Platform-designed exercise Assess one outcome from one randomized trial, recording source locations and reasons for every judgement.
Method boundary RoB 2 assesses randomized-trial results; do not use it as a generic score for interviews or observational studies.
Access Free online chapters: chapter 2 covers scope; chapter 8 covers randomized-trial bias. Open workbook and review criteria →
Source checked 2026-10-25 Evidence synthesis · Handbook and guideline · Advanced · subject foundations
PRISMA statement group · 2020
Reading target Transparent reporting of search, selection and synthesis in systematic reviews.
Prerequisites Distinguish narrative reviews, systematic reviews and meta-analysis.
Platform-designed exercise Populate a flow diagram from your screening log and reconcile deduplication, exclusions and inclusions.
Method boundary A reporting guideline; checklist completion does not establish quality or replace risk-of-bias assessment.
Access Public checklist, explanation and flow-diagram templates. Open workbook and review criteria →
Source checked 2026-10-25 Qualitative interpretation · Handbook and guideline · Advanced · subject foundations
Virginia Braun & Victoria Clarke · 作者维护 / author-maintained
Reading target Distinguish codes, themes and topic summaries, and consider the researcher's interpretive role.
Prerequisites Interview or text material, a question and basic qualitative epistemology.
Platform-designed exercise Build an extract–code–candidate theme–interpretation trail with a counterexample.
Method boundary Do not impose inter-coder agreement as a universal threshold for reflexive thematic analysis; name the analytic approach.
Access Author-site search index available; direct retrieval failed in this check. A worked paper is linked in the guide. Open workbook and review criteria →
Source checked 2026-10-25 Reproducible experiments · Handbook and guideline · Advanced · subject foundations
NeurIPS · 官网现行页 / current official page
Reading target Audit claims, limitations, reproducibility, error reporting and compute resources.
Prerequisites Run a machine-learning baseline independently and understand data splits.
Platform-designed exercise Reproduce a main table with configurations, seeds, cost, variability and discrepancies.
Method boundary A conference checklist does not replace research-method training or a justified experimental design.
Access Public guidance; verify the target year's rules when submitting. Open workbook and review criteria →
Source checked 2026-10-25 Reproducible experiments · Handbook and guideline · Advanced · subject foundations
The Turing Way Community · 持续维护 / living handbook
Reading target Version control, data management, reproducible environments and code documentation.
Prerequisites Basic scripting, command line and Git.
Platform-designed exercise Rebuild a figure from raw data in a clean environment; have another person follow the README.
Method boundary Rerunning a workflow does not establish that its data, design or measurements support the conclusion.
Access Open online handbook. Open workbook and review criteria →
Source checked 2026-10-25 Questions and arguments · Video and course · Advanced · subject foundations
Simon Peyton Jones · 经典讲座 / recorded talk
Reading target Use writing as part of research thinking, organizing around contributions and readers' questions.
Prerequisites An initial finding or a claim that can be tested.
Platform-designed exercise Draft a one-page argument: problem, gap, contribution, evidence, strongest objection and scope.
Method boundary Advice reflects a research tradition; clear writing cannot repair weak evidence.
Access Author page provides video and slides; original video linked below. Open workbook and review criteria →
Source checked 2026-10-25 · Video addresses verified from author pages; recordings were not reviewed end to end. Causal identification · Textbook · Graduate · methods prerequisites
Daniël Lakens · 2022 · 配套在线章节 / companion chapter
Reading target Justify sample size against inferential goals, effects of interest, power, precision and resources.
Prerequisites Effect sizes, confidence intervals and hypothesis testing.
Platform-designed exercise Before seeing main results, specify an effect of interest, sample requirements under alternatives and attrition allowance.
Method boundary Do not mechanically plan from an inflated pilot effect or justify a design with observed post-hoc power.
Access Public author chapter with a companion tool. Open workbook and review criteria →
Source checked 2026-10-25