Techniques

Module 4 · Practice

Units, significant figures, accuracy, precision, and uncertainty

Report measurements with meaningful units and honest digits, while separating closeness to a reference from repeatability.

Units, significant figures, accuracy, precision, and uncertaintyMeasurements can cluster closely yet miss a reference, or scatter around it.precise, biaseddispersed
Field diagramAccuracy and precision
Estimated reading
13 min
Estimated practice
30 min
Equipment
Calculator or spreadsheet; ruler or supplied measurement dataset; no hazardous materials.
Progress stays in this page session. Allow lesson storage in Cookie settings to remember it; browser settings may also prevent saving.

Objective

By the end, you can:

  • Use quantity and unit together and convert without invented precision.
  • Distinguish accuracy, precision, repeatability, and uncertainty.
  • Round and report results so the digits match the measurement process.

A number needs a unit

A measurement is a quantity expressed with a unit and a method. “12” is incomplete: 12 mL, 12 g, and 12 s mean different things. Use a consistent unit system for calculations and label axes, tables, and notebook values. NIST's SI guidance standardizes symbols and advises against carrying more digits through a conversion than the source measurement justifies.

Accuracy describes closeness to a reference or accepted value; precision describes how closely repeated results agree under stated conditions. A set can be precise but biased, or variable but centered near a reference. Uncertainty is a quantified description of doubt in a result based on relevant components such as repeatability, calibration information, resolution, and assumptions. It is not a synonym for error or a promise that the true value lies in a particular interval unless the reporting convention says so.

  • Repeatability: precision under closely repeated conditions.
  • Reproducibility: agreement under changed conditions such as operators or laboratories, when defined by the field.
  • Resolution: the smallest display or scale increment, not automatically the measurement uncertainty.
  • Significant figures: digits supported by the method and data, not a reward for extra decimal places.

Report with discipline

Keep guard digits during intermediate arithmetic, then round the final result using a stated rule. In addition and subtraction, decimal-place precision matters; in multiplication and division, relative precision or significant figures matters. Exact counts and defined conversion factors can be treated differently from measured inputs. Do not report a calculated volume as 10.0000 mL because a formula produced many digits when the original readings support only hundredths.

When comparing measurements, state the reference, method, and uncertainty basis. If no suitable reference exists, describe repeatability or observed spread and say that accuracy was not established. Avoid calling a result “accurate” merely because it looks plausible.

  • Put units in every raw column and carry them through equations.
  • Record the instrument or scale resolution and relevant specification.
  • State whether a spread is range, standard deviation, standard uncertainty, or another defined quantity.
  • Round once at the end unless a method requires a different rule.

Worked example

A qualified statement

The arithmetic mean is 10.03 mL. The readings cluster within 0.02 mL of one another, while the mean is 0.03 mL above the stated reference.

  1. 01Mean = (10.02 + 10.04 + 10.03) / 3 = 10.03 mL.
  2. 02Observed range = 0.02 mL; this is a descriptive spread, not automatically an uncertainty.
  3. 03Relative to the supplied reference, the mean differs by +0.03 mL; reference suitability and method bias remain assumptions.
  4. 04Report: “Three readings gave a mean of 10.03 mL and range 0.02 mL under the stated method.”

Materials

Set out what you need.

  • Calculator
  • Unit-conversion and rounding worksheet
  • Supplied readings with a stated reference for comparison

Safety & stop conditions

Pause if the work no longer fits the plan.

  • No hazardous procedure is required. Measurement and calculation practice should not be used to justify safety-critical or regulated decisions without qualified method review.

Method

Work through the steps.

  1. 01

    Annotate the input

    Write quantity, unit, method, resolution, and reference or calibration information for each value.

  2. 02

    Convert coherently

    Use dimensional cancellation in the equation and retain only digits supported by the source values.

  3. 03

    Compare and qualify

    Compute the chosen summary, identify uncertainty components or limitations, and state whether the claim concerns accuracy or precision.

Checkpoints

Observe, record, investigate.

  • Every number in the result has a unit or is explicitly dimensionless.
  • The reported digits match the least precise relevant input.
  • Accuracy and precision are not used as synonyms.

Common mistakes

Correct the process, preserve the record.

Adding trailing zeros to make a result look more precise.

Report only digits supported by the method and state the resolution or uncertainty basis.

Calling repeated agreement accuracy.

Repeated agreement supports precision; accuracy requires comparison to a suitable reference.

Using “error” to mean uncertainty without defining it.

Name the quantity: observed difference, standard deviation, standard uncertainty, or another defined measure.

Practice

Measure and report

Use supplied readings 10.02, 10.04, and 10.03 mL with a reference of 10.00 mL whose basis is stated as suitable for this exercise. Calculate a mean and describe spread; do not claim more than the data support.

  1. What units and digits belong in the mean?
  2. Does the set show accuracy, precision, or both relative to the reference?
  3. Which uncertainty components are missing from this simple exercise?
Research field kit: worksheets & practice data →

Self-checks

Test your reasoning.

01Can a highly precise set be inaccurate?

Answer: Yes. A systematic offset can make repeated results tightly clustered away from a suitable reference.

02Why should units travel through an equation?

Answer: Dimensional checking can expose a wrong conversion or formula and ensures the final quantity has a meaningful interpretation.

03A calculator shows 10.030000 mL from inputs recorded to 0.01 mL. What should be reported?

Answer: Usually 10.03 mL, with the rounding and method basis stated. The extra display digits are not supported by the recorded inputs.

Sources & further reading

Use the method-specific source when you practice.