NZ Scholarship Psychology common mistakes
NZ Scholarship Psychology · Understand the traps and how to avoid them.
Writing a sequence of study descriptions (aim, method, results) and then moving straight to the next study, so the response reads as a list of summaries rather than an evaluation of evidence.
Internal assessment rewards accurate description, so candidates rehearse study details and carry that habit into a Scholarship essay where the marks sit in appraisal of the evidence, not recall of it.
What to do instead
For every study named, force at least two sentences of appraisal before moving on: what the design lets you claim, what it cannot, and what that means for the argument in the question. Cut the number of studies to three or four so there is room for this.
Presenting both sides of a debate in balanced paragraphs and ending with 'therefore both are important' or no conclusion at all, so the essay has no position that accumulates across paragraphs.
Candidates believe balance signals sophistication and fear that committing to a judgement will be penalised as bias, so they hedge rather than weigh.
What to do instead
State a defensible thesis in the introduction, give each body paragraph a topic sentence that advances it, and reserve one paragraph for the strongest counter-argument plus a reasoned rebuttal. Conclude by stating which evidence carried more weight and why.
Making causal claims from correlational or cross-sectional designs ('this proves that...', 'screen time causes...'), or quoting a precise statistic or p-value that cannot be justified.
Candidates want their argument to sound authoritative, and everyday language about research slides easily into causal phrasing; invented precision feels like evidence of wider reading.
What to do instead
Match verb to design: 'is associated with', 'predicts', 'suggests' for correlational work, and name the alternative explanation (reverse causation, third variable) explicitly. If a figure is not certain, describe the direction and size of the effect qualitatively rather than inventing numbers.
Naming a methodological problem or bias (demand characteristics, social desirability, small sample, experimenter bias) as a label and then moving on, without showing what it does to this particular result.
Textbook lists of 'limitations' are easy to memorise, so they get deployed as tokens of critical thinking rather than as reasoning about a specific finding.
What to do instead
Use a three-step chain every time: the bias exists in this procedure, it would push the measured value in this direction, therefore the claim about X is weakened or survives. If you cannot complete the chain, delete the label.
Describing a theory's components and history when the question asks how well it explains a behaviour or how it compares with a rival account, so evaluation of explanatory power never happens.
Theories are learned as content to be reproduced, and candidates default to the safest material they know under time pressure.
What to do instead
Convert each theory paragraph into a claim about explanatory scope: what the theory predicts that rivals do not, one piece of evidence it handles badly, and one phenomenon it cannot address at all. Use a rival theory as the comparison point throughout rather than describing two theories side by side.
Recommending an intervention in an applied scenario ('the psychologist should use CBT') without stating the mechanism, the target of change, or why it fits this person's presentation.
Case questions invite a quick diagnosis-and-recommendation pattern, and naming a well-known therapy feels sufficient because its mechanism is assumed rather than argued.
What to do instead
Write the fit in three moves: the maintaining process identified in the scenario, the mechanism by which the intervention disrupts it, and the evidence or boundary condition that supports using it here rather than an alternative. Name what you would not choose and why.
Applying findings to a case or to a general claim while ignoring the sample, culture, age, or setting the theory was built on, treating conclusions from one population as universal.
Textbook findings are presented as settled facts, and the qualifying detail about who was studied is easily dropped once the candidate is focused on application.
What to do instead
Before applying a finding, name the population and setting it came from and ask whether the mechanism depends on features that differ in the target case. In a New Zealand context, explicitly consider whether cultural norms, whānau and community structures, or access to services change how the process would operate.
Leaving ethical discussion to a short final paragraph that lists consent, confidentiality and debriefing, disconnected from the study design or recommendation discussed earlier.
Ethics is taught and revised as a separate topic, so candidates treat it as an item to be ticked off rather than a dimension of every design decision.
What to do instead
Integrate ethics at the point of decision: when describing a procedure, note the ethical tension it creates; when proposing a design, explain how you would resolve it and what the resolution costs in validity. A closing paragraph should synthesise a judgement already argued, not introduce ethics for the first time.
Answering ethical questions by reciting a list of principles or code rules as if compliance settles the matter, without acknowledging that principles conflict or making a justified choice between them.
Codes are encountered as rules to be applied, and candidates assume that naming the correct guideline is the same as ethical reasoning.
What to do instead
Identify the specific principles in tension in the scenario (for example, protecting participants from harm against the scientific value of deception), state what each would require, then justify which you would prioritise and under what conditions the other could be accommodated. Show that reasonable psychologists could disagree and answer that disagreement.
Discussing research or practice as culturally neutral, with no reference to the cultural positioning of the researcher, the participants, or the community, and no consideration of who benefits from the knowledge produced.
Candidates read methodology as a set of technical procedures and do not see cultural location as part of rigour, so it never enters the analysis.
What to do instead
Add an explicit cultural lens: whose knowledge is being measured, whether the measure was developed and validated for that group, who was involved in designing the study, and how findings would be returned to the community. Treat this as an evidential point about validity, not a courtesy paragraph.