📘 How do you write a physics lab report?
Methods, figures, and claims—how a lab report becomes something another physicist could actually rerun.
What you’ll learn
- Hypothesis Framing from Prior DataFormulate a physics hypothesis that incorporates quantitative priors and specifies measurable effect sizes.Learners extract tension metrics from published datasets, translate them into a testable prediction with defined statistical power, and anchor the statement to equipment resolution limits. The chapter closes by showing how this framing propagates into every subsequent section of the report.
- Measurement Chain DesignSpecify an end-to-end measurement chain with traceable uncertainty contributions at each stage.Students map sensor-to-analysis signal paths, assign covariance terms, and document auxiliary calibrations required for traceability. Emphasis is placed on writing the methods section so that the chain can be audited without contacting the original team.
- Statistical Model SelectionSelect and justify a statistical model that matches the noise structure and sample size of a physics dataset.The lesson walks through likelihood construction for mixed noise, computation of model evidence, and transparent reporting of selection criteria. Edge cases such as small-n corrections and nested models receive explicit treatment.
- Rigorous Uncertainty PropagationImplement and document uncertainty propagation that satisfies GUM requirements for correlated inputs.Participants derive sensitivity coefficients, run Monte Carlo trials, and format the uncertainty statement so that both methods appear side-by-side with identical coverage factors. Common reporting pitfalls that trigger reviewer rejection are catalogued.
- Figure Construction for RefereesProduce publication-ready figures that preserve quantitative information and meet journal vector requirements.The chapter covers axis scaling, error-bar rendering, color-blind-safe palettes, and caption phrasing that allows a reader to extract numbers without consulting the raw data files. LaTeX and matplotlib workflows are compared for reproducibility.
- Abstract CompressionDraft an abstract that states the central numerical claim while remaining within journal word limits.Learners practice iterative trimming, placement of the key result in the first sentence, and avoidance of hedging language that referees flag. Examples from accepted and rejected abstracts illustrate the difference.
- Methods Section GranularityDetermine the minimum level of experimental detail required for independent reproduction.The lesson supplies decision trees for temperature stability, sample-preparation steps, and software versions. Students rewrite an underspecified methods draft until it meets the standard of a top-tier journal.
- Results Narrative Without OverclaimWrite results text that reports effect sizes and compatibility intervals without implying unsupported causality.Participants translate p-values and credible intervals into journal-appropriate language, insert model-comparison metrics, and flag any post-selection inference. Multiple rewrites of the same dataset demonstrate acceptable versus inflated claims.
- Discussion of Systematic LimitsDocument and quantify the dominant systematic uncertainties that could alter the central claim.The chapter teaches construction of a systematic table, null tests, and cross-checks with independent analysis chains. Emphasis is placed on writing that anticipates the most likely referee objections.
- Conclusion and Outlook FramingCraft a conclusion that summarizes the advance while specifying concrete next experimental steps.Students map the current uncertainty ellipse onto planned facilities or datasets, avoid vague “future work” phrasing, and anchor projections to approved proposals or published roadmaps.
- Reference Curation and AttributionBuild a reference list that satisfies both priority and completeness requirements of physics journals.The lesson covers DOI verification, self-citation limits, and the handling of preprints versus peer-reviewed versions. Tools for automated consistency checking are introduced.
- Reproducibility Package AssemblyAssemble a reproducibility package that meets current journal and funding-agency mandates.Participants create a directory structure containing raw data, processing scripts, environment files, and a manifest that allows bitwise reproduction of every figure and table.
- Response to RefereesDraft a response letter that addresses every technical concern with new data or clarified text.The chapter models the construction of a response matrix, the decision of when to add new figures versus textual clarification, and the tone that avoids antagonizing referees while defending the original analysis.
- Final Submission WorkflowExecute a submission checklist that guarantees consistency across preprint, cover letter, and final files.Learners run a final consistency audit covering equation numbering, figure file versions, ORCID registration, and license selection. The workflow ends with the mechanics of uploading to the journal portal and handling page proofs.
Questions this course answers
Given three priors on Γ with widths 4.2, 3.8 and 5.1 kHz, what inverse-variance weighted uncertainty on the meta-analytic mean should you adopt before choosing Δ?
The inverse-variance weighted uncertainty is the square root of the reciprocal of the sum of the precisions, yielding 2.3 kHz for the given widths.
A new superconducting-resonator paper reports two prior frequency-shift datasets whose central values differ by 1.8 times their combined uncertainty. Which hypothesis statement best converts that tension into a falsifiable claim anchored to a 0.4 MHz resolution floor?
Only the second statement pre-specifies both the quantitative prior location and an effect size large enough to be resolved above the equipment floor while guaranteeing stated power.
Place the calibration links in the correct traceability order from lab instrument to primary standard.
Traceability requires an unbroken sequence from the instrument actually used through transfer standards to a national metrology institute realization; reversing any link breaks the chain.
A new interferometric sensor uses an unstabilized HeNe laser whose frequency drifts 2 MHz per hour. Explain in one sentence why simply quoting the manufacturer’s 0.1 % power uncertainty is insufficient for a traceable budget.
Unstabilized laser frequency introduces a systematic that varies with time and shares a physical origin with the phase observable; omitting its calibration leaves an unquantified covariance that violates traceability.
A new dataset of 850 single-photon counts from a superconducting qubit shows both Poisson statistics and a weak 1/f component. Which information criterion and correction should be used to compare a pure Poisson model against a mixed Poisson-plus-1/f model?
With N = 850 the AICc finite-sample correction is required, and the models are nested so the likelihood-ratio test can be used in tandem.
In your own words, explain why the 2022 Nature Physics coherence-time result changed by 40 % when the analysts switched statistical models.
A likelihood that ignores part of the noise structure yields biased point estimates and incorrect uncertainties; correcting the likelihood changed the inferred coherence time.
Grounded in trusted sources
- National Institute of Standards and Technology
- American Physical Society
- OpenStax
- NIST — reporting measurement results
- American Physical Society — Physical Review style / ethics
- OpenStax University Physics — laboratory practice
Every Wunder lesson is built from real, reputable sources — never invented.
Related courses
Wunder is a personalized learn-anything platform — tell it any topic and it builds a beautiful, fact-checked course in minutes, with narration, a knowledge check, and a college-style University track.
© 2026 Wunder Learning LLC · Terms & Privacy