Prof. Dr. Larry AdamsAcademic, Author & Researcher
Research Methods — Level 7 · Chapter 13 of 20

12. Sampling, Data Collection, and Instrument Design

12.1 Selecting data sources

Primary data are collected for the current study, while secondary data were originally collected for another purpose. Primary data can be tailored to the research question but may require substantial time and ethical review. Secondary data may provide larger or longer-term records but can contain definitions, gaps, or biases that do not match the study. Researchers should check provenance, completeness, comparability, permission, and the implications of using data outside their original context.

12.2 Questionnaire design

A questionnaire should be concise, logically ordered, accessible, and aligned with the constructs in the framework. Questions should avoid double-barrelled wording, leading language, unexplained jargon, and response options that do not cover plausible answers. Sensitive questions need a clear purpose and an appropriate option to decline where possible. Demographic items should be limited to those required for analysis. Pilot testing with people similar to the target population can identify problems before wider distribution.

12.3 Interview guide design

An interview guide should begin with clear, non-threatening questions and progress to more detailed prompts. Questions should invite examples rather than imply the desired answer. “Tell me about a recent project handover” is usually more open than “Why does poor communication always cause delays?” Follow-up prompts can explore sequence, context, decisions, consequences, and alternative interpretations. The interviewer should avoid promising outcomes or implying that a particular response is preferred.

12.4 Access, recruitment, and response bias

Recruitment plans should identify who will invite participants, how eligibility will be checked, what information will be provided, and how consent will be recorded. Response bias may occur when people with strong opinions are more likely to participate, when managers nominate only favourable participants, or when non-respondents differ systematically from respondents. Researchers should document recruitment channels and consider how selection affects conclusions.

12.5 Pilot studies and feasibility

A pilot tests whether procedures work as intended; it is not simply a smaller version of the final study. It can examine recruitment, timing, question comprehension, technology, data quality, and analytic procedures. Pilot findings may require revising an instrument or schedule. If the instrument changes materially, the researcher should document the change and determine whether data gathered under different versions can legitimately be combined.