Market Research and Consumer Behavior
What this domain tests
Research design, sampling, questionnaire quality and consumer behaviour theory: how buyers move from problem recognition to post-purchase, and how involvement changes that path.
Core rules
- Design follows the question: exploratory and qualitative, descriptive via survey or observation, causal via experiment or A/B test.
- Probability samples (simple random, stratified, cluster) support inference to a population; convenience and quota samples do not.
- Margin of error shrinks with the square root of n, so quadrupling sample size halves it.
- Reliability is consistency; validity is measuring the intended construct. A scale can be reliable and still invalid.
- High-involvement purchases follow extended problem solving with search and dissonance reduction; low-involvement repeat buys are habitual, so salience and shelf placement beat argument.
- Decision roles differ: initiator, influencer, decider, buyer, user.
Worked example
In a survey of 400 shoppers, 62% prefer package A. At 95% confidence the margin of error is about 4.9 points, giving an interval of 57.1 to 66.9%, entirely above 50%, so a majority preference is supported. But the response rate was 12%, and non-respondents may differ systematically. Report the result as directional, test the non-response assumption, and only then commit to a packaging change.
Traps
Treating a convenience sample as representative; double-barrelled or leading items; confusing correlation with causation; quoting a point estimate without its interval; ignoring the evoked set, since respondents rate only brands they already know, which inflates awareness-heavy brands.