Skip to content
ProductHow it worksFind a topicProject topics

Business and social science methods Student writing guide

Planning a Business or Social Science Survey and Records Study

A survey is not the default method for every business or social-science question. Existing records, interviews, observation or document analysis may answer a question more directly. If a survey is appropriate, the target population, sampling frame, question wording, recruitment, non-response, measurement and analysis all affect what its answers can support. AAPOR’s survey guidance is a detailed professional reference; adapt it to your local design and institutional rules.

A practical guide from FINAL YEAR NG Check your department's current instructions before you submit.

Decide what evidence the question needs

  1. Specify the population, setting, time period and concept being studied.
  2. Check whether relevant records or reliable public data already exist and whether you are authorized to access them.
  3. Use a survey only when responses from the defined population are needed to answer the question.
  4. Define the sampling frame and how each eligible unit could be selected or invited.
  5. Test question wording for ambiguity, double-barrelled questions and response options before launch.
Worked business-study example

To examine how customers understand a bank’s mobile-app fee notices, a bounded study could compare the wording of approved public notices and invite a defined group of consenting users to complete a short comprehension task. It should not claim that the notices caused account switching unless the design measures that outcome and addresses competing explanations.

Plan the analysis before distribution

Write a question-to-variable plan before sending a questionnaire. Specify how response categories will be coded, how missing answers will be handled, which comparisons are planned and what descriptive output would answer each objective. A large convenience sample does not automatically represent a broader population.

  • Record the population and sampling approach, recruitment mode and field period.
  • Pilot the instrument and revise confusing items before main collection.
  • Track invitations and response counts without storing unnecessary identifiers.
  • Report wording, response options and exclusions so readers can assess measurement.
  • Separate descriptive patterns from causal explanations and note non-response or coverage limits.

Protect private and organisational data

Financial records, student grades, staff evaluations and customer files can contain confidential or identifiable information. Obtain gatekeeper permission and required ethics approval before accessing or recruiting. Use the minimum data needed, restrict access, remove identifiers when permitted and follow retention and deletion requirements.

Sources and further reading

These university writing resources inform the general advice here. They do not replace your department's handbook or supervisor's guidance.

  1. AAPOR: Best Practices for Survey Research
  2. NOUN FMS731 Research Methodology courseware
  3. U.S. Office of Research Integrity: Definition of Research Misconduct
← All writing guidesChoosing a project topic →