Erq practice questions.

A researcher reports that Bandura's (1961) Bobo doll study demonstrates that observing aggression causes children to imitate it. Which is the strongest limitation to raise when evaluating this claim?

• The study used a large representative sample of adults
• The artificial lab setting lowers ecological validity
• Correlational designs cannot establish cause and effect
• The study had no control group for comparison

Answer: The artificial lab setting lowers ecological validity

Bandura used a controlled experiment (so cause-effect is defensible), but the contrived setting and unfamiliar Bobo doll weaken generalisation to everyday aggression — an ecological validity critique.

When evaluating a study that used only male university students from one country, which concept is most directly threatened?

• Internal validity
• Construct validity
• Population validity
• Inter-rater reliability

Answer: Population validity

A narrow, non-representative sample limits how far findings extend to other people — a population validity (external validity) issue, not a flaw in the study's internal logic.

Maguire et al. (2000) found London taxi drivers had larger posterior hippocampi than controls. Which evaluation point is most appropriate?

• Because it was an experiment, we can conclude navigation experience caused the enlargement
• As a correlational study, cause is ambiguous: plasticity or pre-existing differences
• The findings cannot be replicated because MRI is unreliable
• The study proves the hippocampus is only involved in spatial memory

Answer: As a correlational study, cause is ambiguous: plasticity or pre-existing differences

Because participants were not randomly assigned to become taxi drivers, the direction of the relationship is ambiguous — a bidirectional ambiguity typical of correlational/quasi designs.

A study on HM (Milner) is frequently cited as evidence for distinct memory systems. What is the key evaluative strength of such a case study?

• Its findings can be confidently generalised to the whole population
• Rich, in-depth data on a rare condition that cannot be manipulated
• It uses random allocation to control confounding variables
• It produces easily quantified data with high statistical power

Answer: Rich, in-depth data on a rare condition that cannot be manipulated

Case studies of unique patients like HM provide detailed, ethically permissible insight into phenomena that cannot be manipulated, though generalisation from n=1 is limited.

Loftus and Palmer (1974) manipulated the verb in the question ('smashed' vs 'hit'). Why does this design support a causal conclusion about leading questions?

• Because participants were interviewed in their homes
• The IV was manipulated with random allocation
• Because a large naturalistic sample was used
• Because self-reported speed estimates are highly accurate

Answer: The IV was manipulated with random allocation

Controlled manipulation of the wording plus random allocation isolates the IV, allowing a cause-effect claim about how post-event information distorts memory.

Which of the following is the best example of a demand characteristic weakening a study's internal validity?

• Participants guess the aim and change their behaviour to help
• The researcher unconsciously scores the experimental group more favourably
• The sample is drawn only from one culture
• The measuring instrument gives inconsistent results across trials

Answer: Participants guess the aim and change their behaviour to help

Demand characteristics occur when participants infer the hypothesis and alter their behaviour. Unconscious scoring by the researcher is experimenter bias, a one-culture sample limits population validity, and inconsistent measurement is low reliability.