Sri Lanka Scientific Service Exam common mistakes
Sri Lanka Scientific Service Exam · Understand the traps and how to avoid them.
In the General Scientific Knowledge section, spending a fixed number of minutes per MCQ and moving on even when the question is a short calculation, so that easy marks in aptitude are left unattempted or rushed at the end.
Candidates carry over a school-exam habit of equal time per question and do not first scan the paper to classify items by effort. Aptitude items (ratios, unit conversions, graph reading) are often quicker than a single recall item if attempted first, but the candidate meets them late with depleted time.
What to do instead
In the first three minutes, mark every question A (know instantly), B (needs short working), C (needs long working or unsure). Do all A items, then all B, then C. Practise with a timer on mixed sets of 50 items to build the scan-and-sort habit, and keep a hard stop of roughly one minute per item in the first pass.
Treating SI units and dimensional consistency as a formality, so answers are given in mixed units (for example g cm^-3 alongside kg m^-3, or kJ mixed with J) and numerical answers are left without units or with the wrong number of significant figures.
Candidates revise formulas but not the unit system, and under time pressure they copy numbers from the question without converting. Marking schemes in scientific service exams routinely deduct for unit and significant-figure errors even when the arithmetic is right.
What to do instead
Before any calculation, write the target unit next to the answer space and convert all given quantities to SI base units first. Do a dimensional check on the final expression. Do a set of 20 conversion drills (pressure, energy, concentration, molarity to mol m^-3) until conversion is automatic, and state answers to the significant figures justified by the given data.
In the discipline-specific paper, answering from broad textbook memory instead of the specific syllabus sub-topics, so answers are plausible but not the accepted definition, standard method, or standard value used by the discipline.
Candidates prepare from general textbooks and past-paper topics rather than the published syllabus and the standard references of their discipline. Many items test the accepted operational definition, standard test method, or standard constant, where an approximate answer is marked wrong.
What to do instead
Obtain the official syllabus for your subject and map each sub-topic to one standard reference (a recognised textbook or the relevant standard method). For each sub-topic, write a one-page card containing: the formal definition, the standard method or procedure, key numerical constants with units, and two common exceptions. Revise from these cards, not from general reading.
Confusing correlation with causation, or describing a study design that cannot support the stated conclusion, when answering methodology questions or writing the short essay on experimental design.
Candidates learn definitions of study designs in isolation and do not practise matching a design to a research question. Under exam pressure they describe an observational survey and then claim a causal effect, or propose a design without a control or without randomisation.
What to do instead
Practise a fixed template for every design question: research question, exposure and outcome, design type, comparison group, randomisation and blinding, confounders and how they are controlled, and the strongest claim the design permits. Apply the template to at least ten past scenarios and state explicitly whether the design supports association only or causation.
Choosing a statistical test by habit (usually the t-test or chi-square) without checking the data type, the number of groups, pairing, and whether the assumptions hold, and reporting a p-value without an effect size or confidence interval.
Candidates memorise a short list of tests and do not practise the decision path. They also equate statistical significance with practical importance, which is exactly what data-analysis items probe.
What to do instead
Build a one-page decision tree: outcome type (continuous, binary, count, ordinal), number of groups, paired or independent, and assumption checks (normality, equal variance, expected cell counts). For each test, note the non-parametric alternative. In every answer, report the estimate with a confidence interval and an effect size alongside the p-value, and interpret it in the units of the problem.
Misreading or misreporting descriptive statistics and graphs: quoting the mean when the distribution is skewed, confusing standard deviation with standard error, or reading a bar chart of counts as if it showed rates.
Candidates revise inferential statistics and neglect descriptive literacy, yet many aptitude and data-analysis items turn on exactly these distinctions. Standard error is smaller than standard deviation, so a candidate who confuses them reports an implausibly narrow spread.
What to do instead
Drill the distinctions explicitly: mean versus median for skewed data, standard deviation (spread of observations) versus standard error (precision of an estimate), counts versus rates, and absolute versus relative change. For any graph shown, first state what is on each axis and what the denominator is before answering. Practise ten graph-interpretation items per week.
Writing research-methodology answers that omit sampling strategy, sample size justification, and ethical approval, listing only the objective and the instruments.
Candidates treat methodology as a description of what they will do rather than a justification of why each choice is valid. Sampling and ethics are the parts most often skipped in revision because they seem administrative.
What to do instead
Use a checklist for every methodology answer: population and sampling frame, sampling method and why, sample size and its basis, variables and measurement, analysis plan, validity and reliability measures, ethical approval and informed consent, data management, and limitations. Write one full protocol of about 300 words against this checklist and reuse the structure in the exam.
In scientific administration answers, describing the technical solution only and ignoring the administrative instruments: procurement rules, financial regulations, delegation of authority, recruitment procedure, and record-keeping.
Candidates come from a technical background and see administration as common sense, so they answer as a scientist rather than as an officer of a public institution. Public-sector administration items are marked against the formal procedure and the responsible authority, not against what seems efficient.
What to do instead
Learn the actual instruments that govern your workplace: the establishment code, financial regulations, procurement guidelines and their thresholds, and the delegation of authority for your grade. For each, write a short note on the step sequence, the approving authority, and the document trail. Practise scenario answers that name the authority and the step, not just the action.
Answering ethics items by intent ('I would act honestly') rather than by naming the specific principle, the conflicting duties, and the procedure for declaring and managing the conflict.
Candidates treat research and professional ethics as character questions rather than as a body of rules with defined procedures. Items typically test the mechanism (informed consent, ethics review, authorship criteria, data integrity, declaration of interests, whistleblowing channel), not the sentiment.
What to do instead
Prepare a table of core principles (beneficence, non-maleficence, autonomy, justice, integrity, accountability) mapped to concrete procedures: ethics committee approval, consent documentation, authorship criteria, data retention, conflict-of-interest declaration, and misconduct reporting route. For scenario questions, structure the answer as: facts, principles at stake, options, chosen action, and the procedure followed.
Leaving policy questions at the level of general statements about 'development' or 'importance of science', without citing the specific national policy framework, institution, or instrument involved.
Candidates revise science content and neglect the policy context of the scientific service, so they cannot name the framework, the implementing body, or the mechanism by which a policy is executed. Such answers read as filler and score poorly.
What to do instead
Build a one-page map of the national science, technology and innovation policy landscape: the responsible ministry and agencies, the national policy documents, the research and development funding mechanisms, and the technology transfer and intellectual property framework. For each, note one current priority and one implementation instrument. Practise answering policy questions by naming the document, the institution, and the mechanism.