Practical assessment skills

A sound investigation begins with an aim and prediction, identifies the independent and dependent variables, fixes control variables, and selects a safe procedure and suitable apparatus.

Write the prediction in words or as a graph with a stated direction or relationship. Choose a range that can reveal the pattern and a measurement method that responds to the dependent variable.

To test how concentration affects initial rate, vary concentration, measure a time-based rate, hold temperature and total volume constant, and predict how the graph should change.

“Investigate the effect of X” is not a complete aim until the measured outcome, controls and feasible range are specified.

A complete practical plan states the aim, apparatus, labelled arrangement, quantities and concentrations, independent and dependent variables, controls, hazards and how results will be recorded and analysed.

Choose instruments whose range and precision suit the measurement, vary one factor at a time and define how the dependent variable will be calculated. Include enough points to reveal the predicted trend.

For a rate–concentration study, specify a gas syringe or mass-loss method, constant temperature and total volume, a concentration range, repeat runs and a rate calculation from the initial gradient.

A list of apparatus is not a method: it must explain what is changed, what is measured and how the data answer the aim.

Standard quantitative practice reduces avoidable variation: prepare standard solutions accurately, weigh by difference, obtain concordant titres, heat to constant mass and take extra readings near an inflection point.

Each technique solves a different problem. Weighing by difference limits transfer error; constant mass confirms completion; extra points around a sharp change locate an endpoint or transition.

A titration uses a rough titre to locate the endpoint, then concordant titres within the stated tolerance; a gravimetric experiment reheats and reweighs until the mass is constant.

Repeating an inconsistent technique does not create concordance; the method and acceptance criterion must be applied.

Choose the calculation or presentation that matches the question: means for repeats, percentage change or error for comparisons, a gradient for a rate, and a graph with labelled axes and suitable significant figures.

Keep raw and processed data separate, show substitutions and units, and round only at the end. For y = mx + c, interpret m and c in the chemical context rather than reporting numbers alone.

A 10.0% increase from 2.00 to 2.20 is calculated from (0.20/2.00)×100; a rate is the gradient of gas volume against time, not the final volume.

A calculator result is not automatically the correct quantity; check whether the question asks for a mean, percentage, gradient or intercept.

A conclusion should state the observed relationship with quantitative detail, explain it using chemistry, and judge whether the evidence supports the prediction or only a limited claim.

Use the range, scatter, anomalies and uncertainty to qualify the conclusion. A further prediction should follow from the proposed mechanism, and an improvement should target the largest limitation.

“Rate increased as concentration rose over 0.10–0.50 mol dm⁻³; the larger collision frequency explains the trend, but the scatter at 0.50 means a wider claim needs repeats.”

Repeating the hypothesis is not a conclusion; the data and its quality must appear explicitly.

Reliable data are repeatable; valid data measure the intended variable; confidence reflects how strongly the evidence supports the conclusion. Anomalies, narrow ranges, uncontrolled variables and unsuitable apparatus can weaken these separately.

Explain the direction of each weakness: temperature drift may raise or lower rate, poor resolution increases scatter, and measuring the wrong endpoint threatens validity. Replication improves reliability only when the method is valid.

Three concordant titres support repeatability, but a contaminated standard solution can make all of them systematically wrong.

“The results are reliable because they agree” ignores systematic error and validity.