VALID Science 8 common mistakes
VALID Science 8 · Understand the traps and how to avoid them.
When a question says 'explain why the ice melts faster in the metal cup', the candidate names the metal and stops, or restates the observation ('because it is metal') instead of giving a mechanism.
Students treat 'explain' as identical to 'describe' and have not practised the difference between naming a factor and stating the process that links cause to effect.
What to do instead
Before writing, underline the command word and convert it: describe = what happens; explain = why it happens, in a chain of 'because... which means...'. Practise turning each explanation into a two-link causal chain that mentions energy transfer, force, or a named process, and check that the answer would still be true if the object were different.
The candidate reads the method and results but never checks whether the independent variable was the only thing changed; when asked to evaluate the investigation, they say 'it was fair' or 'it was good' without identifying the uncontrolled variable.
Evaluation is taught as a judgement about quality rather than a search for specific confounds; students also assume any school practical is automatically controlled.
What to do instead
Use a fixed audit list on every investigation: What was changed? What was measured? What was kept the same? Then name the single variable that was not controlled and state the direction of its effect on the result. Write the criticism as 'X was not kept constant, so the difference in Y could be due to X rather than the variable tested'.
When asked to predict the reading on a thermometer or the extension of a spring from a table, the candidate writes the nearest table value or the last value rather than interpolating between two rows.
Students read tables as lists of facts to recall rather than as a relationship to be extended; they also lack confidence in simple proportional reasoning.
What to do instead
Practise the routine: identify the pattern between two adjacent rows (difference per unit), state it in words, then apply it to the new value. Always show the arithmetic or the stated pattern in the answer space so partial credit is available, and sanity-check that the prediction lies between the two surrounding table values.
The candidate describes what they would see in a diagram of a food chain or circuit but does not use the correct scientific term, writing 'the animal that eats it' instead of 'predator' or 'consumer', or 'the electricity gets used up' instead of 'energy is transferred'.
Everyday language is fluent and fast, so it crowds out the taught vocabulary under time pressure; students often do not realise the everyday phrase is marked wrong.
What to do instead
Keep a two-column glossary of everyday phrase versus accepted term for each topic and rewrite five everyday sentences into scientific ones each week. In the exam, after writing an answer, circle any word that would appear in a shopping conversation and replace it with the syllabus term.
The candidate draws a results table with no units in the headings, or writes units after every number, and plots a graph with unlabelled axes or an axis that does not start at zero without a break shown.
Presentation rules are learned once and then treated as optional decoration; students focus on getting the numbers and treat axes and headings as an afterthought.
What to do instead
Use a fixed table template: quantity and unit in the heading, plain numbers in the cells. For graphs, apply a four-point check every time: both axes labelled with quantity and unit, even scale, points plotted with small crosses, line or curve of best fit. Rehearse the checklist on three data sets until it is automatic.
The candidate writes that 'the particles expand' when a substance is heated, or says 'heat rises' and 'cold enters', rather than describing particle motion and energy transfer from hot to cold.
Particle diagrams show boxes getting bigger, which invites the literal reading that the particles themselves grow; everyday phrases about heat and cold are absorbed from home language.
What to do instead
Redraw the particle model explicitly: same-size dots, larger spacing, labels 'particles move faster' and 'spacing increases'. Ban the words 'heat rises' and 'cold enters' in written work and require the phrase 'energy is transferred from the hotter region to the cooler region' until it is habitual.
The candidate treats a hypothesis as a guess and writes 'I think the plant will grow taller' with no reference to the variables, and cannot later say whether the data supported it.
Hypothesis is introduced in early years as a prediction, and the link to variables and to a testable statement is never made explicit.
What to do instead
Use the frame 'If [change in independent variable], then [predicted change in dependent variable] because [scientific reason]'. After the data, require a separate sentence: 'The data support/do not support the hypothesis because the trend was...'. Practise on three familiar contexts so the structure, not the content, becomes automatic.
The candidate calculates an average by adding the readings and dividing by the number of readings without discarding an obvious anomaly, or reports the mean to many decimal places that the instrument cannot support.
The mean is learned as a procedure to be executed on all numbers present; students are not taught that anomalous points are excluded and that precision is limited by the instrument.
What to do instead
Before averaging, scan the readings and circle any value far from the rest, then state why it is excluded. Round the mean to the same precision as the raw readings. Write one sentence justifying the exclusion so the reasoning, not just the number, earns credit.
The candidate answers a question about reliability or validity by saying 'repeat it to make it more accurate' without distinguishing repeat readings (reliability) from measuring the right quantity (validity).
The three terms are taught together and sound interchangeable; students memorise one phrase and apply it to every evaluation question.
What to do instead
Learn one-line definitions with a test: reliability = same result on repeat; accuracy = close to true value; validity = the method measures what it claims. For each evaluation question, decide which of the three is at stake first, then choose the matching improvement, and name the specific variable or instrument being changed.
The candidate uses a secondary-source claim such as 'scientists say' or a single trial as evidence, and does not state sample size, control group, or whether the conclusion follows from the data shown.
Students conflate evidence with authority and have little practice judging whether a stated conclusion is actually supported by the numbers given.
What to do instead
Apply a three-question evidence check to any claim: How many subjects or trials? Was there a control or comparison? Does the stated conclusion match the size and direction of the difference? Write the judgement as 'the conclusion is/is not supported because the difference was X with Y trials and no control was described'.