Collecting Data

Objectives and outcomes

• Objective - what the study wants to find out

  “Does taking multivitamins improve well-being?”

• Outcomes - how the answer is measured

  • Primary - main result e.g. pain relief)

• Secondary - other info (e.g. side effects)

Surrogate outcomes

• short cues used instead of waiting for full clinical results

• Faster and easier to measure

E.g. surrogate outcome is blood pressure and the real outcome is heart attack/stroke

• Don’t always guarantee the real outcome will improve

Types of data

Numerical

• Continuous (e.g. blood pressure)

• Discrete (e.g. number of siblings)

Categorical

• Nominal (e.g. blood type)

• Ordinal (e.g. pain scale 1–5)

Data collection

Clinical Data

• Collected with tools/instruments (e.g. blood tests, BP monitor)

• More objective

Self-Reported Data

• Collected by asking the patient (e.g. diet diaries, surveys)

• Easy to collect but can have recall or response bias

Interviews

• Standardised: Same questions, same way

• Non-standardised: More open or flexible

• Useful for attitudes or personal experiences

Questionnaires

• Collect demographics and behaviours - compare to other populations

• Formats: paper, online, face-to-face

• Use validated tools for better accuracy

Types of questions

• Open: “What are your thoughts?” (detailed answers)

• Closed: “Yes/No” or multiple choice

• Double-barrelled: Asking two things at once

• Leading questions: Push for a certain answer

Scales for responses

• Likert scale - scale (1-5, agree-diagree)

• Semantic differential - (painful - painless)

• visual/numeric scales - for pain, mood

Reliability and validity

• Reliability = consistency

  • Do you get the same result if repeated?

  • Types:

    • Inter-rater: Do different people get the same result?

    • Test-retest: Same result at different times?

• Validity = accuracy

  • Does the tool measure what it’s meant to