Grading Rubric - Modern Physics Laboratory

How the laboratory grade is composed¶

Component	Weight	When
Pre-lab questions	10%	Collected at the start of each period, 14 times
Notebook	20%	Initialled weekly; graded in Weeks 3, 6, 10, 14
Short reports (11)	35%	One week after the period
Full reports (3)	25%	Weeks 4, 10, 14 — two weeks after the period
Participation and technique	10%	Continuous

The lowest short-report score is dropped. Full reports are not dropped.

Report rubric¶

Each report is scored out of 100 against the criteria below. The weighting is deliberate: the discussion of uncertainty is worth more than the result. A report with a 12% discrepancy that correctly identifies why, and bounds it, outscores one with a 1% discrepancy and no error analysis.

Report scoring

Criterion	Pts	What full credit looks like
Abstract	5	Three to four sentences. States the method and the numerical result with its uncertainty and unit. A reader learns the answer without reading on.
Theory	15	The working equation is derived or cited to a specific chapter section, every symbol is defined, and the assumptions that could fail in this apparatus are named. No unnecessary reproduction of the textbook.
Apparatus and method	10	A labeled diagram or annotated photograph. Model numbers, controlled quantities, and ranges. A competent reader could repeat the measurement.
Data and figures	15	Data plots have labeled axes with units and points with error bars; fitted models appear as smooth curves. Every figure has a caption stating what to notice, including diagrams and photographs. Tables carry units and uncertainties. Nothing is plotted outside the data’s range.
Analysis	20	The fit is weighted by real uncertainties. Parameter uncertainties come from the covariance matrix. $\chi^2_\nu$ is reported and interpreted. Residuals are shown or discussed. Propagation to the final quantity is correct and shown.
Discussion of uncertainty	20	Type A and Type B contributions are separated and quantified. The dominant term is identified, with a concrete statement of what would reduce it. Systematic effects are named with a sign or a bound. The comparison to the accepted value is made in units of $\sigma$ , not percent alone.
Conclusion and writing	10	One paragraph with the number and what it means. Prose is clear, tenses consistent, significant figures correct throughout. References present.
Notebook and code	5	Notebook pages for this experiment are attached or scanned; the Jupyter notebook runs top to bottom and produces the numbers in the text.

Automatic deductions¶

These are applied on top of the criteria above, because they are habits worth breaking early:

−5 per figure with an unlabeled axis or a missing unit.
−5 per result quoted without an uncertainty.
−5 for uncertainties quoted to more than two significant figures, or values carried to more digits than the uncertainty supports.
−10 for an unweighted fit to logarithmically transformed data (see Uncertainty and Curve Fitting).
−10 for a discussion whose only named error source is “human error”.
Up to −100 for data or text that is not your own. Partners share data; everything else is individual work. See the course syllabus.

Notebook rubric¶

Graded four times a semester, 25 points each.

Notebook scoring

Criterion	Pts	What full credit looks like
Contemporaneity	6	Entries were plainly made at the bench: ink, in order, dated, with mistakes struck through and left legible rather than erased or recopied.
Completeness	6	Date, partner, apparatus identity, a labeled sketch with dimensions, raw readings in the instrument’s own units, and settings.
Uncertainty at the point of measurement	5	Each reading carries the uncertainty you assigned at the time, with a one-line justification of where it came from.
Narrative of what went wrong	4	Failed attempts, adjustments, and the reasoning behind them are recorded. A notebook with no problems in it is either luck or fiction.
Running conclusion	4	A sentence at the end of each period stating the result obtained and whether it looked reasonable.

Pre-lab rubric¶

Each pre-lab is scored out of 10, collected at the start of the period, and not accepted late — its purpose is to prepare you for the next three hours, which it cannot do afterwards.

4 — All questions attempted, with work shown.
4 — The order-of-magnitude estimate for the headline quantity is correct and is the number you will compare your first reading against.
2 — A one-sentence statement of what you expect to see, specific enough to be wrong.

Participation and technique¶

Assessed continuously, not by attendance alone:

Arrives having read the experiment, not having skimmed it.
Aligns and operates the apparatus rather than watching a partner do it, and swaps roles at the halfway point.
Handles lasers, sources, and instruments according to Laboratory Safety without needing to be reminded.
Leaves the bench in the state they would want to find it.
Asks questions that follow an attempt to answer them.