PSLE RME common mistakes
PSLE RME · Understand the traps and how to avoid them.
When a word problem gives two quantities and asks for the total, the candidate adds even when the relationship calls for subtraction, multiplication, or division, choosing the operation from the numbers rather than from the question.
Candidates scan for numbers and keywords and act on surface cues instead of building a picture of what is being asked; many have memorised 'total means add' without checking whether the question asks for a part, a difference, or a rate.
What to do instead
Train a three-step routine before any calculation: underline the question sentence, restate it in the candidate's own words as a sentence with a blank for the answer, then name the operation and justify it in one line. Practise with paired problems that use identical numbers but different questions so the operation must change.
The candidate writes the final answer only, with no working, so method marks are lost even when the final value is right, and errors cannot be traced.
Candidates believe only the answer is marked, or they run out of time and skip working to save space; some do rough work in margins or on scrap and never transfer it.
What to do instead
Set a classroom rule that every problem starts with a written number sentence or diagram, and mark practice papers using a scheme that awards method marks. Show a worked script where the answer is wrong but most marks are earned, so the payoff is visible.
Units, decimals, and remainders are dropped or mismatched: an answer of 3.5 chairs, a length given in metres when the question used centimetres, or a remainder ignored when the question asks how many full boxes are needed.
Candidates treat the number as the answer and stop; they do not re-read the question after computing, and they have not been taught to interpret remainders in context.
What to do instead
Require a final checking line: write the answer with its unit, then ask 'does this make sense in the story?' Use a short set of remainder-interpretation tasks (round up for buses, ignore for leftover money) and a conversion drill where the unit is part of the marked answer.
In comprehension, the candidate lifts a whole sentence from the passage, or copies a phrase that contains the topic word but does not answer the question actually asked.
Candidates equate finding the topic with answering the question; they lack a strategy for identifying the question type (cause, purpose, feeling, inference) and so default to copying.
What to do instead
Teach question-type labelling first: for each question, decide whether it asks for a reason, a description, an inference, or the meaning of a word, then locate the sentence that supplies that specific information. Require answers phrased in the candidate's own words and mark lifting as incomplete.
Grammar and tense errors inside otherwise correct answers: subject-verb agreement slips, wrong tense in a reported-speech or summary task, and pronouns with no clear referent.
Candidates compose the content of the answer first and never proofread the sentence; under time pressure they edit meaning but not form, and they have not internalised agreement rules well enough to apply them while writing.
What to do instead
Build a five-item personal editing checklist (verb agrees with subject, tense consistent, pronoun clear, sentence complete, punctuation correct) and enforce a two-minute proofread at the end of every practice paper. Use short daily sentence-correction exercises drawn from the candidate's own scripts.
In composition, the candidate writes a long build-up and then ends abruptly with no resolution, or the story drifts away from the given topic or picture.
Candidates plan the opening in detail because that is what they have practised, and they run out of time or ideas before the climax; some do not check the prompt's requirements before starting.
What to do instead
Teach a fixed planning grid (characters, setting, problem, two attempts, resolution, closing reflection) completed in three minutes before writing, with a time budget that reserves the last five minutes for the ending. Practise writing only the climax and resolution for many different prompts.
The candidate misreads a Setswana instruction or question word, answering a 'why' question with a 'what', or confusing near-identical forms such as the negative and affirmative of the same verb.
Instruction vocabulary and question words are met mostly in reading, not in structured vocabulary work, so candidates decode the passage but not the task; interference from English question patterns also plays a part.
What to do instead
Run a weekly question-word drill in Setswana (what, who, where, when, why, how, how many) matched to short passages, requiring the answer type to be named before writing. Keep a class wall chart of instruction verbs with worked examples.
Noun-class agreement breaks down in the candidate's own Setswana sentences: the subject concord, adjective concord, and pronoun do not match the noun class of the subject, especially with plural and human-class nouns.
There are many classes and concords to hold in memory, and candidates apply the concord of the last noun they saw rather than the subject; errors in speech are tolerated, so the habit carries into writing.
What to do instead
Use concord tables organised by noun class with a model noun in each, and drill sentence-building where the subject noun is swapped while the rest of the sentence must be adjusted. Mark concord errors separately from content errors so the pattern becomes visible.
When reporting results, the candidate describes what happened ('the water went up the straw') instead of stating the conclusion or explaining why, and does not link the observation to the scientific idea.
Candidates are not shown the difference between an observation, a pattern, and an explanation; they believe a vivid description is a complete answer, and explanation language has not been modelled.
What to do instead
Model three-column tables (what I saw / what it means / why it happens) and require every practical write-up to include one 'because' sentence. Give partial-credit practice where a description alone scores less than a description plus explanation.
In fair-test questions, the candidate changes more than one variable, or names the variable to keep constant as something unrelated to the investigation.
Candidates focus on making the experiment 'work' rather than on making it a fair comparison, and the vocabulary of variables is new; they also assume any sensible-sounding control is acceptable.
What to do instead
Use a standard planning frame for every investigation: question, what I will change, what I will measure, what I will keep the same, and what I predict. Have candidates critique flawed plans written by others and justify each correction in one sentence.