Marketing Research and Information Systems
Marketing research and information systems covers how firms gather, analyze, and use data to reduce uncertainty. Questions test the research process, secondary versus primary data, research designs, sampling, questionnaire quality, and information-system use.
The research process
- Define the problem and objectives
- Develop the research plan
- Collect data
- Analyze data
- Present findings
- Secondary data already exist and should be checked first; they are cheaper and faster but may be outdated or not fit the problem.
- Primary data are collected for the specific problem.
- Exploratory research clarifies a problem; descriptive research measures or profiles; causal research tests cause and effect, often with experiments.
Sampling and measurement
- Probability samples: simple random, systematic, stratified, cluster. Nonprobability samples: convenience, judgment, quota, snowball.
- A larger sample reduces random error but does not fix a biased sampling frame.
- Questionnaires should avoid leading, double-barreled, and ambiguous questions; pretesting catches problems.
- Reliability means consistency; validity means measuring what is intended.
Information systems and ethics
- A marketing information system is people, equipment, and procedures that gather, sort, analyze, evaluate, and distribute information.
- Analytics and CRM data support segmentation and measurement; privacy, consent, and security are ethical requirements.
Worked example
A retail chain sees declining store traffic. Exploratory focus groups suggest signage and hours matter. A descriptive survey measures how often customers notice signage and when they shop. A causal experiment changes hours in matched stores and compares traffic with control stores. The chain uses a stratified sample by store region, not a convenience sample of loyalty members, because the latter excludes nonmembers. Reasoning: match design to the question, use secondary data first, and protect representativeness.
Traps that cost marks
- Starting primary research before checking secondary data.
- Confusing a sample with the population.
- Believing a bigger sample corrects a biased frame.
- Confusing correlation with causation.
- Accepting a leading question because it produces the desired answer.
- Treating reliability and validity as synonyms.