Lebanese University Faculty of Medical Sciences Entrance Examination - Second Year, General Medicine common mistakes
Lebanese University Faculty of Medical Sciences Entrance Examination - Second Year, General Medicine · Understand the traps and how to avoid them.
When converting between units in stoichiometry problems (e.g., mg to mol, mL to L, or g/L to mol/L), the candidate substitutes the number into the formula without first writing every quantity in base SI units, then reports an answer off by 10^3 or 10^6.
Lebanese secondary and first-year teaching often mixes conventional units (mg, mL, °C, atmosphere) with SI units, and candidates memorize formulas like n = m/M or c = n/V without internalizing that M must be g/mol and V must be in litres. Under time pressure they plug in whatever number appears in the prompt and skip the conversion step entirely.
What to do instead
Adopt a fixed three-line ritual before any calculation: (1) list each given quantity with its unit, (2) convert every quantity to base SI in a separate written line, (3) only then substitute. Practice with a set of 20 mixed-unit drills where the only task is conversion, not solving, so the habit becomes automatic before it is tested inside a larger problem.
The candidate identifies the limiting reagent by comparing the given masses or moles of the two reactants directly, rather than dividing each by its stoichiometric coefficient, and then computes the product yield from the wrong reactant.
Textbook examples frequently use reactants in a 1:1 ratio, so the shortcut 'smaller number of moles wins' appears to work. When the exam presents a 2:3 or 1:3 stoichiometry, the candidate carries the shortcut over and the error propagates into yield and excess calculations. The underlying concept - that the coefficient is part of the mole ratio - is often memorized as a label rather than understood.
What to do instead
Drill the rule 'divide moles by coefficient, smallest quotient is limiting' on at least ten problems with non-1:1 coefficients, and require the candidate to write the quotient for each reactant explicitly before proceeding. Include at least three problems where the intuitively 'smaller' mass is not the limiting reagent, to break the mass-based heuristic.
On organic reaction questions the candidate predicts the major product by pattern-matching the functional group name alone (e.g., 'alkene + HBr → Markovnikov product') without checking whether the substrate is allylic, benzylic, conjugated, or subject to rearrangement, and without checking for stereochemical outcomes.
Reaction prediction is taught as a catalogue of named rules, and exam prompts often use the simplest substrate, so surface matching is rewarded in practice. Real exam items deliberately use substituted or conjugated systems where the simple rule gives the minor product. Candidates also tend to ignore stereochemistry because they treat it as a labelling exercise rather than a consequence of mechanism.
What to do instead
Force mechanism-first reasoning: before naming a product, draw the intermediate carbocation or transition state, mark its stability (degree, resonance, hyperconjugation), and only then apply the rule. Build a one-page decision tree for each reaction class (electrophilic addition, substitution, elimination) that includes the rearrangement and stereochemistry branch, and practise on substrates that trigger each branch.
In genetics problems the candidate confuses the observed phenotypic ratio with the expected ratio and treats a deviation as proof of linkage or epistasis, without performing or interpreting a chi-square test or checking whether the sample size could explain the deviation by chance.
Mendelian ratios (9:3:3:1, 3:1, 1:1) are memorized as facts, and candidates learn to recognize them by eye. When a ratio does not match, the default explanation in their mental model is 'linkage', because that is the next chapter. The statistical reasoning that a small sample produces noisy ratios is rarely practised, and chi-square is often learned as a formula rather than a decision rule.
What to do instead
Require a written chi-square calculation with explicit degrees of freedom and a stated critical value before any genetic interpretation. Practise with deliberately small samples where the deviation is not significant, so the candidate learns to accept the null hypothesis. Separately, practise distinguishing linkage from epistasis by asking whether the deviation affects all classes or merges specific classes.
When asked to trace a pedigree or calculate a probability of inheritance, the candidate assigns genotypes by guessing from affected individuals alone, does not mark carriers, and does not propagate the probability through the pedigree generation by generation.
Pedigree analysis is often taught by recognition of the inheritance pattern (autosomal recessive, X-linked, mitochondrial) rather than by systematic genotype assignment. Candidates also treat probability as a single multiplication and forget that conditional information ('unaffected sibling of an affected child') changes the prior probability, a step that requires Bayes-style updating.
What to do instead
Impose a fixed procedure: label every individual with all possible genotypes, eliminate possibilities using each affected individual, then compute probabilities forward from the founders using the product rule and, where relevant, condition on known phenotypes. Practise at least five pedigrees that require conditional probability, and require the candidate to state the prior and the condition explicitly in writing.
In physics problems the candidate substitutes numbers into a memorized equation before drawing a free-body diagram or defining the coordinate system, so sign errors and missing forces (normal force, friction, tension, component of gravity) appear, and the answer is dimensionally plausible but physically wrong.
Physics is perceived as a formula-recall subject, and candidates who are strong at algebra are rewarded for fast substitution. Free-body diagrams are often treated as an optional illustration rather than the computational scaffold. Without a defined positive direction, signs are assigned ad hoc, and forces that 'seem obvious' are omitted under time pressure.
What to do instead
Make the free-body diagram mandatory and graded: every force drawn with a label, every axis drawn with a positive direction, and components resolved before any equation is written. Practise a set of inclined-plane and multi-body problems where the only task is to produce the diagram and the component equations, with no numerical answer required.
The candidate answers a data-interpretation item by reading the single highest or lowest point on a graph or table and generalizing from it, ignoring trend, axis scale, units, sample size, and whether the comparison is controlled.
Data interpretation is often practised as 'find the number in the table', which rewards point-reading over reasoning. Candidates also skip axis labels because the figure looks familiar, and they conflate correlation with causation when the prompt describes an observational study. Under exam time pressure, the first salient number becomes the answer.
What to do instead
Train a fixed four-question protocol for every figure: what is on each axis and in what units, what is the overall trend, what is the sample size or error bar, and what claim does the data actually support versus merely suggest. Practise on items where the correct answer is 'the data do not support this conclusion' or 'the difference is within the error bars', so the candidate learns to reject overreach.
When a question uses a technical term whose everyday meaning differs from its scientific meaning (e.g., 'theory', 'significant', 'control', 'affinity', 'specificity', 'diffusion'), the candidate answers from the everyday sense and does not notice the mismatch.
Scientific vocabulary is acquired partly from popular media, where 'theory' means a guess and 'significant' means important. The candidate reads the stem fluently and never flags the term as a technical one, so the conceptual error is invisible to them. This is especially common for candidates whose scientific reading in the language of instruction is limited to textbooks and who have not seen the terms in varied contexts.
What to do instead
Maintain a personal glossary of 'false-friend' terms with a one-line scientific definition and a contrasting everyday example, and review it weekly. In practice, require the candidate to underline every technical term in the stem and restate its scientific definition before choosing an option. Add targeted items where the everyday reading produces a plausible but wrong distractor.
In cell and molecular biology items the candidate describes a process as a static list of components (e.g., 'DNA replication involves helicase, polymerase, ligase') without ordering the steps, specifying directionality (5'→3'), compartment (nucleus versus cytoplasm), or the regulatory condition under which the process occurs.
Revision is often done by memorizing vocabulary lists and diagrams as labels, which supports recall questions but fails on questions that ask what happens if an enzyme is inhibited, in which direction synthesis proceeds, or where in the cell a step occurs. The candidate has the nouns but not the causal chain, so any option containing the right nouns looks correct.
What to do instead
Rehearse each core process as a numbered sequence with direction, compartment, and a 'what breaks if this step fails' line. Use self-testing that asks for the consequence of a specific block (e.g., 'polymerase inhibited: what accumulates, what fails, where?') rather than for definitions. Draw the process from memory weekly and compare against the source.
On anatomy and physiology items the candidate recalls the name and location of a structure but not its function, its blood supply or innervation, or its relationship to neighbouring structures, so questions framed around function, lesion, or spatial relation are answered by name-matching.
Anatomy is frequently revised through labelled diagrams and lists, which build recognition memory for names and positions. Physiology is revised separately, so the integration between structure and function is never practised. Exam items that ask 'which structure is at risk in this lesion' or 'what happens if this nerve is damaged' therefore defeat a purely nominal knowledge base.
What to do instead
Study each region as a functional unit: for every structure, write its function, its arterial supply, its innervation, and one neighbouring structure it could be confused with. Practise lesion-based questions ('damage here produces which deficit?') and spatial questions ('what lies immediately medial/lateral/anterior?') rather than label-the-diagram questions.