RULE Dual Degree Economics Entrance common mistakes
RULE Dual Degree Economics Entrance · Understand the traps and how to avoid them.
In optimisation problems, the candidate writes the Lagrangian, sets every partial derivative to zero, and then treats the multiplier equations as the solution, never reporting the optimal quantities or checking second-order conditions.
The Lagrangian is the most mechanical part of the problem, so it feels like the answer. Candidates have often practised 'set up and solve the FOCs' drills without ever being graded on the final economic statement or on whether the stationary point is a maximum.
What to do instead
Adopt a fixed four-line closing routine for every optimisation answer: (1) solve the FOC system explicitly for the choice variables in terms of parameters, (2) state the multiplier and interpret it as the shadow price of the constraint, (3) verify the second-order condition (bordered Hessian for constrained problems, or concavity/quasi-concavity argument), (4) write one sentence of economic interpretation. Practise until the closing routine costs under two minutes.
The candidate computes a comparative-static derivative such as dx*/dp and reports its sign from a single numerical example rather than from the general expression, or reports the derivative without translating its sign into an economic statement.
Numerical examples are faster and less error-prone than algebra, so candidates substitute numbers early and then believe the resulting sign is general. Many have never been explicitly taught that a comparative-static claim is a statement about all admissible parameter values.
What to do instead
Force generality: derive the sign analytically using the implicit function theorem or the envelope theorem, and only then check with one example as a sanity test. Write the sign conclusion in words, e.g. 'demand for the good falls as its price rises because the substitution effect dominates here, since the good is a small budget share'. If the sign is ambiguous, state the condition under which each sign occurs.
For an IS-LM or AD-AS question, the candidate shifts one curve, reads off the new equilibrium, and stops, ignoring the feedback to the other market, the labour market, or the assumed price flexibility.
Textbook diagrams are usually drawn one shock at a time, and candidates memorise 'increase in G shifts IS right' as an isolated fact. Under exam time pressure, the second-round effect is the first thing dropped.
What to do instead
Before writing, state the model's closure in one line: which prices are sticky, which market clears, what the policy rule is. Then trace the shock through every market in sequence and mark each curve that moves. Finish by contrasting the answer under the alternative closure (fixed vs flexible prices) in two sentences, because that contrast is usually where the marks sit.
The candidate defines a macroeconomic identity incorrectly, for example writing that national saving equals investment in an open economy, or confusing the government budget deficit with the trade deficit without stating the accounting condition.
Identities are memorised as slogans rather than derived from the expenditure decomposition, and the open-economy version is studied quickly and then forgotten.
What to do instead
Derive the identity once from Y = C + I + G + NX and rearrange to S = I + NX (private plus public saving). Keep that derivation on a single revision card and re-derive it at the start of any question involving twin deficits, capital flows, or the current account. Never quote an identity you cannot reproduce in three lines.
When asked to compute a derivative, integral, or limit, the candidate applies the rule mechanically without checking the domain, differentiability, or whether an improper integral converges, and presents a divergent or undefined object as a number.
Technique drills reward speed on well-behaved functions, so existence conditions are treated as boilerplate. Candidates rarely see marks deducted for this in school-level practice.
What to do instead
Add a two-second domain audit to every calculus answer: state the domain of the function, note any point where the rule fails, and for improper integrals evaluate the limit explicitly. If the limit does not exist or the integral diverges, say so in words and explain why, since a correct 'diverges' answer scores better than a fabricated finite value.
The candidate solves a system of linear equations by elimination but never checks whether the coefficient matrix is invertible, so a singular or near-singular system produces a confident but meaningless unique solution.
Matrix methods are taught as procedures for solving, not as tests of existence. In economic applications, singularity often carries meaning (redundant equilibrium conditions), which candidates do not expect.
What to do instead
Compute the determinant (or rank) before or alongside solving, and interpret: det = 0 means no unique solution, so state whether the system is inconsistent or has infinitely many solutions. In an economic context, connect singularity to the economic structure, for example two goods being perfect substitutes so relative prices are indeterminate.
The candidate reports a regression coefficient as a causal effect, or interprets a p-value as the probability that the null hypothesis is true.
Software output invites a mechanical reading of the coefficient table, and the p-value misinterpretation is extremely widespread in secondary teaching. Candidates often have no practice writing the correct sentence.
What to do instead
Write interpretations from fixed templates. For a coefficient: 'a one-unit increase in X is associated with a b estimate of Y, holding the other regressors fixed; this is causal only if the zero conditional mean assumption holds, which requires exogeneity of X'. For a p-value: 'if the null were true, we would see a statistic at least this extreme with probability p'. Then add one sentence on what would threaten the causal claim (omitted variable, simultaneity, measurement error).
The candidate answers a statistics question about a sampling distribution or confidence interval using the logic of the population distribution, for example saying the probability that the true mean lies in a given 95% interval is 0.95.
The frequentist distinction between the fixed parameter and the random interval is subtle and is often taught as a formula rather than as a probability statement, so the intuition defaults to the familiar population picture.
What to do instead
Memorise the correct phrasing and use it verbatim: 'across repeated samples, 95% of intervals constructed this way would contain the true parameter; this particular interval either does or does not'. Pair it with a one-line contrast to the Bayesian credible interval so the distinction sticks, and check every interval answer against that phrasing before moving on.
In an analytical reasoning set, the candidate reads the stimulus once, begins answering, and then re-reads the passage for every option, so time runs out and later questions are guessed.
Candidates treat reasoning sections like knowledge tests, where the passage is a lookup source, rather than like puzzles where the structure must be extracted first. Anxiety about time encourages starting immediately.
What to do instead
Spend the first 30 to 45 seconds diagramming: conclusion, premises, and the exact gap between them, in symbols or arrows. Then predict the answer before looking at the options, and eliminate rather than confirm. For logic games, build the full set of constraints and one or two hypothetical arrangements before answering any question.
In the essay or extended written answer, the candidate describes the topic and lists both sides without committing to a thesis, and uses hedging phrases ('it could be argued', 'many factors are involved') in place of an argument.
School writing often rewards balanced coverage and length, whereas this exam rewards a defensible position supported by evidence and explicit reasoning. Candidates transfer the wrong genre conventions.
What to do instead
Draft a one-sentence thesis containing a claim and a because-clause before writing anything else, and open the essay with it. Structure each body paragraph as claim, evidence, and link back to the thesis, and reserve hedging for genuine uncertainty only. In the final two minutes, check that every paragraph either supports, qualifies, or answers the thesis, and cut any paragraph that only describes.