Collecting Data
- Design survey and interview instruments that avoid common measurement errors
- Distinguish leading, double-barreled, and neutral questions
- Select data-collection procedures that fit the method and protect data quality
Instruments turn questions into data
A data-collection instrument — a survey, an interview protocol, an observation guide, a coding sheet — is the tool that operationalizes your research question. Its design determines the quality of everything that follows: no analysis can rescue data from a badly built instrument. For surveys, that means well-worded items and appropriate response scales; for interviews, a protocol of open questions and planned follow-ups; for observation, a clear, consistent scheme for what counts as what. Pilot-testing an instrument on a few people before full collection catches confusing items early.
How question wording distorts data
Poorly worded questions bias responses before analysis begins. A leading question ("Don’t you agree the cafeteria food is terrible?") pushes respondents toward an answer. A double-barreled question ("Is the food tasty and affordable?") asks two things at once, so an answer is uninterpretable. Loaded or emotionally charged wording, vague terms ("often" — how often?), and unbalanced scales all skew results. Good items are neutral, ask one thing, use concrete language, and offer balanced response options. In interviews, the parallel is asking open, non-leading questions and truly listening rather than steering.
Procedures protect data quality
Beyond the instrument, the procedure — how, when, and where you collect — affects quality. Standardize conditions so every participant has a comparable experience (same instructions, same setting where possible). For interviews, record and transcribe accurately so you analyze what was actually said, not your memory of it. Keep a data-management plan: label, back up, and secure your data. Consistency here supports reliability, and careful, uniform procedures reduce the chance that an artifact of your process — not the phenomenon — drives your findings.
Critique and rewrite this survey item measuring student attitudes: "Wouldn’t you say our overcrowded, underfunded school needs more modern technology and better teachers?"
- 1.Spot the leading framing: "Wouldn’t you say" and the loaded adjectives "overcrowded, underfunded" push the respondent toward agreement.
- 2.Spot the double- (really triple-) barreled structure: it bundles technology and teachers into one question, so a single answer cannot be interpreted.
- 3.Spot the emotional/loaded wording that biases the response before the person even considers it.
- 4.Rewrite as separate, neutral items with balanced scales — e.g., "How satisfied are you with the technology available in your classes?" (Very dissatisfied → Very satisfied).
- 5.Add a second, separate item for teachers, similarly neutral and balanced.
A survey asks, "How satisfied are you with the school’s food and its price?" What is the main flaw?
Pilot-test your instrument on three to five people before full data collection. You will catch confusing wording, double-barreled items, and broken scales while they are still cheap to fix — after collection, they are permanent flaws in your data.
Why should qualitative interviewers record and transcribe interviews rather than rely on notes from memory?
Reviewers scrutinize alignment between your instrument and your question. Include or describe your actual survey items or interview protocol in an appendix, and justify your wording choices — a hidden or poorly worded instrument undermines the credibility of every result you report.
Answer the 2 checkpoints as you read.
Sign in to save your progress