Sitonce
Country: US
Show exams for United States Hong Kong
Sign in

Measurement, Significant Figures and Experimental Error (FTCE 003)

Updated 10 min read
Key takeaway

A measurement consists of a numerical value, unit, and uncertainty determined by the method or instrument.

  • Writing “12” without a unit or context is incomplete.
  • Read the scale at its appropriate resolution and record estimated digits only when the procedure calls for them.
On this page21 sections
  1. Measurements include units and uncertainty
  2. Accuracy, precision, and validity
  3. Systematic and random error
  4. Significant figures
  5. Dimensional analysis
  6. Density and specific heat
  7. Graphs and linear relationships
  8. Uncertainty and comparison
  9. Worked example: density
  10. Worked example: percent error
  11. Laboratory records and reproducibility
  12. Common errors and practice
  13. Calibration and blanks
  14. Precision in repeated trials
  15. Percent difference and percent error
  16. Significant figures for logarithms
  17. Use dimensional checks before arithmetic
  18. How to review this topic
  19. Additional reasoning and applications
  20. Apply the idea carefully
  21. Repeatability in a chemistry investigation

Measurements include units and uncertainty

A measurement consists of a numerical value, unit, and uncertainty determined by the method or instrument. Writing “12” without a unit or context is incomplete. Read the scale at its appropriate resolution and record estimated digits only when the procedure calls for them. Repeated values may vary because of instrument resolution, procedure, and genuine sample differences. Avoid reporting more precision than the instrument supports. A calculated result cannot be more informative than its inputs, though retaining guard digits during intermediate steps reduces rounding error.

Accuracy, precision, and validity

Accuracy concerns closeness to an accepted reference value; precision concerns agreement among repeated measurements. Measurements can be precise but inaccurate if a balance is miscalibrated, or accurate on average but imprecise if readings vary widely. Validity asks whether the method measures the intended quantity and tests the stated question. Calibration, consistent technique, appropriate controls, and representative sampling contribute to stronger evidence. Identify the problem before suggesting a fix: more repeated readings may improve precision but will not correct a systematic calibration error.

Systematic and random error

Random error produces scatter that can vary in direction across measurements. Repeated trials and suitable statistical summaries help characterize it. Systematic error shifts results in a consistent direction, such as a zeroed instrument with an offset. Repeating the same faulty procedure can produce very precise wrong results. A blunder, such as misreading a scale or recording the wrong sample, is not a normal uncertainty source and should be investigated. Do not use “human error” as a complete explanation; name the specific mechanism and its expected effect.

Significant figures

Significant figures communicate measurement precision. In multiplication and division, the final result is commonly rounded to the fewest significant figures among measured inputs. In addition and subtraction, decimal-place limits govern. Exact counts and defined conversions do not limit precision in the same way as measured values. Scientific notation makes significant digits explicit. These conventions are useful reporting rules, not substitutes for a full uncertainty analysis. Carry guard digits through intermediate steps and round once at the end to avoid cumulative error.

Dimensional analysis

Dimensional analysis uses conversion factors so units cancel along a calculation pathway. For example, converting milliliters to liters uses one liter per one thousand milliliters; arrange the fraction so milliliters cancel. In a chemistry problem, units can reveal whether a mole ratio, molar mass, or concentration conversion belongs next. If units do not cancel to the requested quantity, the setup is wrong even if the arithmetic appears neat. Include units at each line and use a final magnitude check.

Density and specific heat

Density is mass divided by volume, an intensive property under specified conditions. Specific heat relates heat transfer to sample mass and temperature change. These calculations depend on unit consistency and an appropriate material assumption. If a measured density differs from a reference, possible explanations include temperature, composition, volume-reading error, or calibration. Do not identify contamination as the only cause without evidence. A careful interpretation separates the measured value from proposed explanations.

Graphs and linear relationships

A graph can reveal proportionality, trends, and outliers. Read axes and units before fitting a relationship. In a straight-line model y = mx + b, slope m gives change in y per unit x and intercept b is the predicted y when x is zero, if meaningful. A line through the origin supports direct proportionality within the measured range. Do not extrapolate far beyond the data without justification. Scatter around a line reflects residual variation, not automatically a flawed experiment.

Uncertainty and comparison

When two results include uncertainty intervals, apparent numerical differences may not establish a meaningful difference. The interval type matters: range, standard deviation, standard error, and confidence interval answer different questions. If the prompt does not define error bars, do not assume what they represent. Report the difference and the uncertainty available. A result that overlaps a reference value may be compatible with it, but formal inference needs the design and statistical assumptions. Keep interpretation proportionate to precision.

Worked example: density

A solid sample has mass 27.4 g and volume 10.2 mL. Density is mass divided by volume, approximately 2.69 g/mL, reported to three significant figures because both measurements have three. If the cylinder uncertainty is large, the last digit may overstate practical certainty despite the rule. If the sample is irregular, displacement volume must avoid trapped bubbles and account for meniscus reading. The arithmetic is only one piece; measurement procedure determines whether the result is trustworthy.

Worked example: percent error

Percent error is often calculated as the absolute difference between an experimental and accepted value, divided by the accepted value, times 100 percent. The accepted value is the denominator because the comparison asks how large the deviation is relative to the reference. This does not identify the cause of the difference. If results all lie above the reference, investigate possible systematic bias; if values scatter on both sides, random variation may contribute. Check the problem’s stated convention and units before calculating.

Laboratory records and reproducibility

A useful lab record states the sample identity, procedure, instrument, units, environmental conditions, and observations needed to reproduce the work. Record unexpected events rather than silently correcting them. Preserve raw data separately from processed results. If a method changes midway, document when and why. Reproducibility does not mean every lab obtains identical values; it means another qualified person can follow the method and understand the uncertainty. This practice makes evidence auditable and helps distinguish chemical or sample variation from procedural differences.

Common errors and practice

Common mistakes include omitting units, rounding too early, mixing exact and measured quantities, confusing accuracy with precision, and claiming a single cause for a discrepancy. Practice one significant-figure calculation, one unit conversion, one graph slope, and one error interpretation. For each, state what the result means and what it cannot establish. Check if the final unit answers the question and if the numerical precision matches the measurements. This habit catches many errors before they become a broader chemical interpretation.

Calibration and blanks

Calibration compares an instrument’s response with known standards across a relevant range. A blank contains the components of a measurement system except the analyte, helping identify background signal. A calibration curve relates signal to concentration and should not be extrapolated far beyond its measured standards without support. A blank does not correct every possible error. Document instrument settings, standards, and units so another analyst can assess whether the method was appropriate for the sample.

Precision in repeated trials

Repeated trials help estimate random variation and improve the stability of a mean when the measurements are independent and the protocol is consistent. Standard deviation describes spread among observations; standard error estimates uncertainty in a sample mean under assumptions. These are not interchangeable. A mean can obscure a bimodal distribution or outliers, so inspect the data shape. Report sample size and the summary measure used. A larger number of technical readings from one preparation does not replace independent samples.

Percent difference and percent error

Percent error typically compares one measurement with a known accepted reference and uses the reference in the denominator. Percent difference compares two measurements when neither is treated as the accepted standard and may use their average in the denominator, depending on the stated convention. Textbooks can define the terms differently, so follow the formula supplied by the question. Never swap the denominator without explanation. Both quantities summarize relative discrepancy; neither identifies why the values differ.

Significant figures for logarithms

For logarithmic values such as pH, common chemistry conventions connect the number of decimal places in the logarithm to significant figures in the original concentration. A concentration with two significant figures is typically reported with two decimal places in pH. Because logarithms compress large ranges, check whether the concentration is plausible for the reported pH. Exact reporting rules can vary by problem instruction, so prioritize explicit directions and carry adequate digits in intermediate work.

Use dimensional checks before arithmetic

Dimensional checking can catch errors in formulas as well as conversions. Density must reduce to mass per volume; molarity must reduce to moles per liter; specific heat calculations should yield energy. If a formula produces units inconsistent with the desired property, inspect the equation before calculating. Unit cancellation also helps choose the correct conversion-factor orientation. Treat units as part of the algebra, not as decoration added after the number is found.

How to review this topic

Use the official competency outline to identify which actions are expected, then practice explaining each process in your own words. Work at least one example with units and one unfamiliar scenario. Keep a short error log that names the reasoning step you missed, such as choosing the wrong model, skipping a conversion, or overclaiming from limited evidence. Revisit the same skill later with changed values or conditions. Retrieval and explanation make it easier to transfer knowledge when the test presents a new context.

Additional reasoning and applications

A calibration example shows why measurement traceability matters. If absorbance is measured for standards of known concentration, a graph can establish an empirical relationship between signal and concentration. An unknown’s signal is then compared with that curve within its calibrated range. A blank can account for solvent or reagent background, but it will not fix a dirty cuvette, mislabeled sample, or detector drift. Replicate measurements of a standard can reveal precision, and periodic checks can show whether calibration remains stable. A slope with units of signal per concentration helps verify how a concentration is inferred. If the line is nonlinear, the analyst should use a validated nonlinear model or a narrower range rather than force a straight line. The calibration procedure should match sample matrix and conditions when those factors affect signal. In exam scenarios, identify the standard, blank, unknown, and measured response. Then distinguish a measurement issue from a chemical explanation. A consistent offset across standards suggests a systematic issue; a scattered set may reflect random error or inconsistent technique. More precise instrument readings do not automatically mean that the method is accurate or valid.

Apply the idea carefully

When comparing measurements from groups, inspect the distribution and collection process before using a mean. If one sample is an outlier, ask whether it reflects a documented procedural failure or legitimate chemistry. Predefine an exclusion rule when possible; otherwise report the value and explain a sensitivity check. A data display should show all relevant observations where practical, not only a summary that hides variation. Describe the size of an effect in original units and use a relative measure when it helps compare scales.

Repeatability in a chemistry investigation

A repeatable result depends on a clearly stated procedure, consistent instrument setup, and independent measurements where appropriate. Repeating a titration with the same buret technique can show how endpoint judgment varies. Measuring one sample several times estimates measurement consistency but does not represent variation among different samples. Preserve raw readings, record calibration and environmental conditions, and report the summary statistic used. If precision is poor, improve technique or measurement resolution; if all values are biased in one direction, investigate calibration or method assumptions. These diagnoses point to different corrections.

Common questions

How should I approach percent difference and percent error?

Percent error typically compares one measurement with a known accepted reference and uses the reference in the denominator. Percent difference compares two measurements when neither is treated as the accepted standard and may use their average in the denominator, depending on the stated convention. Textbooks can define the terms differently, so follow the formula supplied by the question. Never swap the denominator without explanation. Both quantities summarize relative discrepancy; neither identifies why the values differ.

How should I approach significant figures for logarithms?

For logarithmic values such as pH, common chemistry conventions connect the number of decimal places in the logarithm to significant figures in the original concentration. A concentration with two significant figures is typically reported with two decimal places in pH. Because logarithms compress large ranges, check whether the concentration is plausible for the reported pH. Exact reporting rules can vary by problem instruction, so prioritize explicit directions and carry adequate digits in intermediate work.