Skills

Physics Experiment Execution and Analysis

10 weeks · 0 milestones

Design and execute a real physics experiment: state a specific testable prediction from physical theory, collect raw data with documented methodology and measurement uncertainty, perform error analysis (systematic and random errors, error propagation), and compare results to theoretical predictions with a quantitative assessment of agreement. The proof is the complete experimental record including raw data, error analysis, and a written conclusion discussing the comparison to theory. For students without lab access, analysis of a named real physics dataset from CERN Open Data Portal, NASA Exoplanet Archive, SDSS, or LIGO Open Science Center constitutes an equally valid proof — the methodology documentation requirements (data provenance, analysis steps, uncertainty quantification, comparison to theory or published values) are identical to lab work. Reviewed by a physicist who examines the error analysis methodology and asks how the results would change if a specific systematic error were introduced.

Part of: Natural Sciences · Proves for: UG

Milestone map

3 milestones

Design experiment and specify measurement protocol

1 week

Design a specific physics experiment, identify what physical quantity you will measure, define the measurement protocol, and select either a real physical setup or a computational simulation. Accessible alternative: if physical laboratory equipment is unavailable, a well-designed computational simulation using free tools (Python/scipy, GNU Octave, PhET simulations) can produce equivalent experimental evidence — the data analysis and error treatment are identical for both routes. Real experiments using household equipment (optics with a laser pointer, mechanics with a pendulum, thermodynamics with a water bath) are encouraged when feasible.

Proof required

Submit your experimental design document: the physical quantity being measured, the theoretical relationship being tested, the equipment or simulation tool being used, the measurement protocol (what you will vary, what you will keep constant, how many measurements at each condition), and your uncertainty budget identifying the main sources of measurement error.

What gets checked

Physical quantity is specific and measurable — not 'investigate how pendulums work' but 'measure the dependence of pendulum period on length and determine g with uncertainty'
Theoretical relationship is stated before the experiment — the expected mathematical relationship between the variables must be stated in advance so that the data can test it rather than fit it
Uncertainty budget identifies at least three distinct sources of measurement error — random measurement uncertainty, systematic instrument uncertainty, and at least one physical source (e.g., air resistance in a pendulum, fringe detection in optics)

Common mistakes

Not stating the theoretical prediction before data collection — a physics experiment is designed to test a prediction, not to describe whatever data emerges; the prediction must come first
Not considering the range over which the theoretical relationship holds — simple relationships often hold only within a limited range; the experimental design must consider where the approximation used in the theory breaks down

Resources

Foundationstart here

Depthgo deeper

What a verifier looks for

Ask the submitter to state the theoretical prediction they are testing — confirms they have framed the experiment as a test, not a description.
Ask what the main systematic error is and how they will estimate it — distinguishing random from systematic errors is fundamental experimental physics.
Ask what the expected result is and what would count as agreement with theory — a quantitative comparison criterion must be defined before data collection.

You'll sign in first, then come straight back here.

Execute experiment, collect data, and perform error analysis

1–3 weeks

Execute the experiment (real or simulated), collect the measurements, and perform a full error analysis. Error analysis is as important as the measurements themselves in experimental physics — a result reported without uncertainty is not a scientific result. The error analysis must include both the propagated uncertainty on the final derived quantity and a check for systematic errors.

Proof required

Submit your raw data table (all measurements with individual uncertainty estimates), your error propagation calculation showing how the uncertainty on each measurement propagates to the uncertainty on the final derived quantity, and your comparison of the experimental result (with uncertainty) against the theoretical prediction — stating whether the result is consistent with theory within the quoted uncertainty.

What gets checked

Raw data is submitted (not only the derived result) — all individual measurements with their assigned individual uncertainties
Error propagation is shown explicitly — not 'the uncertainty is ±X' but the calculation showing how each measurement's uncertainty contributes to the final derived uncertainty
Result is compared against theory with a quantitative consistency statement — 'the measured value of g = 9.6 ± 0.3 m/s² is consistent with the accepted value of 9.81 m/s² at the 1σ level' rather than 'close to the accepted value'

Common mistakes

Reporting only the final derived quantity without the error propagation — a result without uncertainty is not a scientific measurement
Performing too few measurements to detect systematic effects — a single series of measurements cannot distinguish random from systematic errors; at minimum two independent measurement sets should be compared

Powstik Guide

A raw data table with uncertainties and a full error analysis.

Steps

Record every reading with its uncertainty.
Repeat readings where you can.
Propagate uncertainty to the final result.
Name the biggest error source.

Template

| Reading | Value | Uncertainty |
|---------|-------|-------------|
Final result: ___ ± ___
Propagation working:
Largest error source:

What gets sent back

Readings without uncertainties.
No propagation shown.
Data edited without a note.

Resources

Foundationstart here

Depthgo deeper

What a verifier looks for

Ask the submitter to walk through the error propagation for the main measured quantity step by step — confirms they computed it rather than estimated it.
Ask whether the result is consistent with the theoretical prediction at a specific confidence level — tests whether they understand what 'consistent within uncertainty' means quantitatively.
Ask whether there is evidence of a systematic error in the data — for example, whether repeated measurements drift in one direction, suggesting a time-dependent systematic effect.

You'll sign in first, then come straight back here.

Write lab report and present with Q&A

1–2 weeks to write and schedule review

Complete a physics laboratory report in the standard scientific format and present to a physicist for a Q&A that probes the experimental design, error analysis, and physical interpretation of the results. A lab report in the physics tradition must be written at a level of detail that allows the experiment to be reproduced by a competent physicist from the report alone.

Proof required

Submit your complete lab report (2000–3500 words in standard format: abstract, introduction with theoretical context, experimental methods with full equipment/simulation specification, results with tables and figures, error analysis, and discussion with comparison to theory and sources of systematic error) plus a Q&A record showing specific challenges to the experimental approach or data interpretation and your responses. The reviewer must be named and their experimental physics background stated.

What gets checked

Methods section is reproducible — all experimental parameters, equipment specifications, and software tools named with version; a physicist should be able to reproduce the experiment from this section alone
Discussion addresses discrepancies between the measured and theoretical values — even small discrepancies must be explained; 'close enough' is not a scientific discussion of agreement
Q&A record shows at least two substantive challenges and responses

Common mistakes

A discussion that declares agreement with theory without addressing the discrepancy quantitatively — 'our result agrees with theory' when the measured value differs by 5% requires an explanation of whether 5% is within uncertainty and what might explain it
A reviewer without experimental physics experience — the Q&A must probe data quality and error analysis, which requires experimental expertise

Powstik Guide

A full lab report, defended to a physicist.

Steps

Build an error budget for every measured quantity.
Compare your result to theory with numbers.
Say whether the gap is within uncertainty.
Name the likely systematic errors.
Present to a physicist and log their challenges.

Template

| Quantity | Value | Uncertainty | Source of error |
|----------|-------|-------------|-----------------|
Result vs theory: ___ vs ___ (difference ___σ)
Systematic errors:

Challenge session log
Date:            Duration:
Reviewer name:
Reviewer role and credentials:

Challenge 1 (error analysis or interpretation):
  What they asked:
  My answer:
  Did it hold?
Challenge 2:
  What they asked:
  My answer:
  Did it hold?
Challenge 3:
  What they asked:
  My answer:
  Did it hold?

Update I am committing to:

What gets sent back

'Agrees with theory' with no numbers.
A reviewer without experimental physics experience.
No systematic errors discussed.

Resources

Foundationstart here

Depthgo deeper

What a verifier looks for

Ask the submitter to explain the physical interpretation of the discrepancy between the measured and theoretical values — this is where experimental understanding is demonstrated.
Ask what the main source of systematic error was and how they estimated its magnitude — quantitative error estimation is the core experimental skill.
Ask what they would change if they ran the experiment again — tests whether they understood the limitations of their design.

You'll sign in first, then come straight back here.

Part of

Physics & Applied Mathematics →

Related outcomes

Milestone map

3 milestones

Design experiment and specify measurement protocol

1 week

Proof required

What gets checked

Physical quantity is specific and measurable — not 'investigate how pendulums work' but 'measure the dependence of pendulum period on length and determine g with uncertainty'
Theoretical relationship is stated before the experiment — the expected mathematical relationship between the variables must be stated in advance so that the data can test it rather than fit it
Uncertainty budget identifies at least three distinct sources of measurement error — random measurement uncertainty, systematic instrument uncertainty, and at least one physical source (e.g., air resistance in a pendulum, fringe detection in optics)

Common mistakes

Not stating the theoretical prediction before data collection — a physics experiment is designed to test a prediction, not to describe whatever data emerges; the prediction must come first
Not considering the range over which the theoretical relationship holds — simple relationships often hold only within a limited range; the experimental design must consider where the approximation used in the theory breaks down

Resources

Foundationstart here

Depthgo deeper

What a verifier looks for

Ask the submitter to state the theoretical prediction they are testing — confirms they have framed the experiment as a test, not a description.
Ask what the main systematic error is and how they will estimate it — distinguishing random from systematic errors is fundamental experimental physics.
Ask what the expected result is and what would count as agreement with theory — a quantitative comparison criterion must be defined before data collection.

You'll sign in first, then come straight back here.

Execute experiment, collect data, and perform error analysis

1–3 weeks

Proof required

What gets checked

Raw data is submitted (not only the derived result) — all individual measurements with their assigned individual uncertainties
Error propagation is shown explicitly — not 'the uncertainty is ±X' but the calculation showing how each measurement's uncertainty contributes to the final derived uncertainty
Result is compared against theory with a quantitative consistency statement — 'the measured value of g = 9.6 ± 0.3 m/s² is consistent with the accepted value of 9.81 m/s² at the 1σ level' rather than 'close to the accepted value'

Common mistakes

Reporting only the final derived quantity without the error propagation — a result without uncertainty is not a scientific measurement
Performing too few measurements to detect systematic effects — a single series of measurements cannot distinguish random from systematic errors; at minimum two independent measurement sets should be compared

Powstik Guide

A raw data table with uncertainties and a full error analysis.

Steps

Record every reading with its uncertainty.
Repeat readings where you can.
Propagate uncertainty to the final result.
Name the biggest error source.

Template

| Reading | Value | Uncertainty |
|---------|-------|-------------|
Final result: ___ ± ___
Propagation working:
Largest error source:

What gets sent back

Readings without uncertainties.
No propagation shown.
Data edited without a note.

Resources

Foundationstart here

Depthgo deeper

What a verifier looks for

Ask the submitter to walk through the error propagation for the main measured quantity step by step — confirms they computed it rather than estimated it.
Ask whether the result is consistent with the theoretical prediction at a specific confidence level — tests whether they understand what 'consistent within uncertainty' means quantitatively.
Ask whether there is evidence of a systematic error in the data — for example, whether repeated measurements drift in one direction, suggesting a time-dependent systematic effect.

You'll sign in first, then come straight back here.

Write lab report and present with Q&A

1–2 weeks to write and schedule review

Proof required

What gets checked

Methods section is reproducible — all experimental parameters, equipment specifications, and software tools named with version; a physicist should be able to reproduce the experiment from this section alone
Discussion addresses discrepancies between the measured and theoretical values — even small discrepancies must be explained; 'close enough' is not a scientific discussion of agreement
Q&A record shows at least two substantive challenges and responses

Common mistakes

A discussion that declares agreement with theory without addressing the discrepancy quantitatively — 'our result agrees with theory' when the measured value differs by 5% requires an explanation of whether 5% is within uncertainty and what might explain it
A reviewer without experimental physics experience — the Q&A must probe data quality and error analysis, which requires experimental expertise

Powstik Guide

A full lab report, defended to a physicist.

Steps

Build an error budget for every measured quantity.
Compare your result to theory with numbers.
Say whether the gap is within uncertainty.
Name the likely systematic errors.
Present to a physicist and log their challenges.

Template

| Quantity | Value | Uncertainty | Source of error |
|----------|-------|-------------|-----------------|
Result vs theory: ___ vs ___ (difference ___σ)
Systematic errors:

Challenge session log
Date:            Duration:
Reviewer name:
Reviewer role and credentials:

Challenge 1 (error analysis or interpretation):
  What they asked:
  My answer:
  Did it hold?
Challenge 2:
  What they asked:
  My answer:
  Did it hold?
Challenge 3:
  What they asked:
  My answer:
  Did it hold?

Update I am committing to:

What gets sent back

'Agrees with theory' with no numbers.
A reviewer without experimental physics experience.
No systematic errors discussed.

Resources

Foundationstart here

Depthgo deeper

What a verifier looks for

Ask the submitter to explain the physical interpretation of the discrepancy between the measured and theoretical values — this is where experimental understanding is demonstrated.
Ask what the main source of systematic error was and how they estimated its magnitude — quantitative error estimation is the core experimental skill.
Ask what they would change if they ran the experiment again — tests whether they understood the limitations of their design.

You'll sign in first, then come straight back here.