Where students lose marks

The pattern

High-mark calculations unravel when cubic centimetres are used as cubic decimetres, actual and theoretical product masses swap places in percentage yield, or the whole product mass is used for atom economy.

What to do instead

Write the governing equation and convert units before substituting. Label the numerator and denominator in words, then round only the final answer to the requested precision.

The pattern

Planning marks disappear into vague phrases such as 'same amount' or 'human error'. Students also identify the material rather than the property deliberately changed or controlled.

What to do instead

Name a measurable quantity with its unit, describe how it changes or stays constant, and tie each improvement to a specific source of error and its effect on the result.

The pattern

Methods commonly begin with the product instead of the named reactants, mix reaction and separation steps, or evaporate all the water rather than allowing crystals to form.

What to do instead

Separate the method into reaction, removal of excess solid, concentration, cooling, filtration and drying. Check that the chosen acid and base can actually make the requested salt.

The pattern

Answers may mention electrons yet omit that they are delocalised and able to move through the structure. Some descriptions imply random collisions rather than a current produced by directed movement.

What to do instead

State the bonding model, identify the mobile charged particles and explain how their movement transfers charge through the metal when a potential difference is applied.

The pattern

A rising straight line is often described as increasing acceleration even when its constant gradient shows otherwise. Instantaneous-speed questions also attract point-reading instead of a tangent construction.

What to do instead

Say what each axis and gradient represent before interpreting the shape. For an instantaneous value, draw a tangent, choose two well-separated points on it and include units.

The pattern

Frequency methods sometimes count waves without measuring a time, while speed investigations alter frequency instead of the intended physical variable. Apparatus names alone do not explain how a value is obtained.

What to do instead

Specify what is counted, the timed interval, repeat strategy and calculation. When testing a relationship, vary the stated independent variable and describe how each dependent reading is produced.

The pattern

Current and resistance rules are frequently transferred to the wrong circuit arrangement. Equation mixing then compounds the error, even when each individual formula is known.

What to do instead

Mark junctions first, decide which quantity is shared or split, reduce the network one section at a time, and use standard circuit symbols throughout.

The pattern

Generator answers drift into motor descriptions, and transformer explanations claim that changing turns directly changes power. References to magnet movement often omit the changing field through the conductor.

What to do instead

Name the changing magnetic flux, the induced potential difference and the resulting alternating current where appropriate. For transformers, compare the turns and potential-difference ratios before discussing losses.

The pattern

Percentage work often becomes a chain of disconnected additions, particularly for repeated or reverse changes. A plausible final number can hide that the original quantity was never represented.

What to do instead

Write each change as a multiplier and keep the unknown original as a variable. Apply the complete multiplier chain, then solve rather than undoing percentages by subtraction.

The pattern

Speed, density, pressure and area-volume problems lose marks when conversions are delayed or squared and cubed units are treated as linear. Students may reach a number with no interpretable unit.

What to do instead

Convert every dimension before using the formula, write the compound unit beside each intermediate quantity, and sense-check whether the scale of the answer fits the context.

The pattern

Multi-step problems often jump straight to arithmetic, leaving no equation that connects the stated relationships. Any later value can then be mistaken for the quantity the question actually requests.

What to do instead

Define the unknown, translate each condition into an expression or equation, solve with visible algebra, and finish by interpreting the solution in the original setting.

The pattern

Data comparisons frequently quote a median or range without saying which group is larger or more consistent. Comments about graph shape can replace the numerical comparison the marks require.

What to do instead

Make one contextual comparison of central tendency and one of spread, each using values from the graph or table and a clear comparative word.

The pattern

Connected-particle and rough-surface solutions often omit a force, assign an action-reaction pair to one body, or assume friction equals its limiting value without justification.

What to do instead

Isolate each body, label every force and choose positive directions. Use friction as an inequality unless limiting equilibrium or impending motion makes equality valid.

The pattern

Statics methods can use a labelled length instead of the perpendicular distance to a force's line of action. Reactions are also included unnecessarily when a better pivot would eliminate them.

What to do instead

Choose the pivot strategically, sketch each line of action and mark its shortest distance from the pivot before writing clockwise and anticlockwise moments.

The pattern

Probability products are sometimes written as though events are independent when the context does not establish that. Conditional information is then applied to the wrong denominator or branch.

What to do instead

Define the events, decide whether they are mutually exclusive or independent, and write the relevant conditional or product rule before substituting values.

The pattern

Normal-distribution work loses direction when a verbal condition is converted to the wrong tail, or when a sum or mean is treated as a single observation with unchanged variance.

What to do instead

Write the event as an inequality, identify the distribution of the required random variable, standardise with its own mean and variance, then check the answer against a sketch.

The pattern

Differential-equation solutions can stop after the complementary function, reuse a trial form that clashes with it, or apply initial conditions before the full general solution exists.

What to do instead

Classify the auxiliary roots, write the complementary function, choose an independent particular-integral form, combine them, and only then use the supplied conditions.

The pattern

Impact models are spoiled by applying restitution to a tangential component or conserving momentum for one sphere alone. Signs may also change between the momentum and separation equations.

What to do instead

Choose normal and tangential directions, keep one sign convention, apply impulse and restitution only where the contact force acts, then test that the bodies separate after impact.

The pattern

Work-energy arguments often mix external work, potential-energy change and kinetic-energy change without a declared system. Incorrect signs can survive several lines before producing an impossible speed.

What to do instead

State the bodies in the system and the zero level, mark work done on or by the system, and audit the final energy balance for sign and physical feasibility.

The pattern

Hypothesis-test solutions can find a probability but never define the rejection region or state a contextual conclusion. Type II error and power are also treated as fixed labels rather than parameter-dependent probabilities.

What to do instead

Write both hypotheses, construct the critical region to meet the stated size, compare the observation, and conclude in context. Evaluate Type II error under the specified alternative, then subtract from one for power.

The pattern

Students may differentiate successfully yet stop at a stationary x-value, set the second derivative to zero, or omit the maximum area or volume named in the question.

What to do instead

Set the first derivative to zero, solve, justify the nature of the stationary point, and substitute back to evaluate the quantity that the question asks for.

The pattern

A common route finds the critical values but reports only one boundary, reverses the accepted interval, or forgets to count the integer values satisfying a contextual restriction.

What to do instead

Find every critical value, test intervals or use a sign diagram, intersect with physical restrictions, and translate the final interval into the requested values or count.

The pattern

Answers often give the calculator's principal angle only, lose a negative square root, or omit valid quadrants within the stated domain.

What to do instead

Find the reference angle, use the sign of the trig function to choose all valid quadrants, and list only solutions inside the supplied interval.

The pattern

Proofs can contain correct angle values but rely on an unstated symmetry, assume a quadrilateral is cyclic, or leave angle statements unlabelled. The numerical result alone does not establish the argument.

What to do instead

Name each angle with three letters, state the exact circle or parallel-line theorem that licenses it, and build a continuous chain from the given facts to the conclusion.

The pattern

Long calculations lose their chemical meaning when mole ratios, equilibrium amounts and partial pressures are mixed in one line. A correct equilibrium idea can still fail through an inconsistent table or unit.

What to do instead

Write the balanced equation, label the mole or concentration change, build the equilibrium row, and convert mole fractions to partial pressures before entering a Kp expression.

The pattern

Students may draw a secant where an instantaneous tangent is needed, infer an order from raw rate changes without holding other concentrations steady, or treat ln A as A in Arrhenius work.

What to do instead

Mark the exact comparison being used, calculate concentration factors and rate factors separately, and let the units of k check the total order before using the rate equation.

The pattern

Mechanisms are weakened by arrows that start in empty space, incomplete bond breaking, or a stability explanation that names a carbocation without the electron-donating inductive effect. Stereochemical consequences may then be asserted without the required geometry.

What to do instead

Start each arrow at a bond or lone pair and finish at the electron destination. After drawing the intermediate, explain its relative stability and use planar or chiral geometry only where the structure supports it.

The pattern

Transition-metal answers can treat counterions as ligands, give a remembered colour for the wrong ligand environment, or explain a colourless ion without checking its d-electron count.

What to do instead

Identify the central ion's oxidation state, coordination number and actual ligands first. Then write the substitution or redox equation and attach the observation to that specific species.

The pattern

Practical answers often say that a method is 'more accurate' without connecting a resolution or scatter to the measured quantity. Gradient limits may be drawn too narrowly or truncated before the uncertainty is calculated.

What to do instead

State the dominant measurement uncertainty, use well-separated points for extreme acceptable gradients, and express the final absolute or percentage uncertainty with matching precision.

The pattern

The negative gradient is sometimes reported as the resistance without handling its sign, while emf and terminal potential difference are combined as unrelated quantities. Circuit suitability is then guessed rather than justified.

What to do instead

Begin from the terminal-potential equation, map its intercept and gradient to physical quantities, and state how the chosen circuit changes current before comparing power or voltage.

The pattern

Gas calculations fail when Celsius enters a proportional argument, amount in moles is confused with molecule count, or pressure and volume are compared separately when their product is required.

What to do instead

Convert temperature to kelvin immediately, decide whether the equation needs amount or particle number, and write the complete before-and-after relationship before cancelling constants.

The pattern

Definitions lose marks by omitting the test charge, mixing vector field strength with scalar potential, or treating charging and discharging exponentials as interchangeable. Equipotential geometry is sometimes drawn without the required perpendicular relationship.

What to do instead

Anchor each definition to force, energy and unit charge as appropriate. Sketch the field and equipotentials first, then choose the capacitor equation whose initial and final values match the physical process.