preparation of data collection and questionnaire
Asking for Preferences
Asking for Preferences
Here is a thorough guide on data collection methods and questionnaire preparation for a mixed-methods clinical study.
Data Collection and Questionnaire Preparation in Mixed-Methods Clinical Research
1. What Is Mixed-Methods Research?
Mixed-methods research combines quantitative data (numbers, statistics, measurable outcomes) with qualitative data (narratives, experiences, perceptions). In a clinical study, this typically means pairing a structured questionnaire or registry with in-depth interviews or focus groups.
The three most common designs:
| Design | How It Works | Example |
|---|---|---|
| Sequential Explanatory | Quantitative first, then qualitative explains the numbers | Survey patients, then interview a subset about surprising results |
| Sequential Exploratory | Qualitative first, then quantitative validates themes | Focus groups shape the questionnaire items |
| Concurrent/Triangulation | Both collected simultaneously, then merged | Survey + interviews run in parallel, compared at analysis |
2. Preparing Your Data Collection Plan
Before writing a single question, work through these steps:
Step 1 - Define Research Objectives Clearly
- State your primary research question (what you want to measure or understand).
- Identify sub-questions: what specific variables, outcomes, or experiences matter?
- Distinguish which questions are best answered by numbers (prevalence, severity scores, treatment outcomes) and which need narrative (patient experience, barriers, decision-making).
Step 2 - Identify Your Population and Sampling Strategy
| Quantitative Component | Qualitative Component |
|---|---|
| Probability sampling (random, stratified, systematic) for generalizability | Purposive or theoretical sampling for rich, relevant perspectives |
| Calculate sample size using power analysis | Sample until data saturation (typically 10-20 participants) |
In a clinical study, common populations include: inpatients, outpatients, caregivers, clinicians, or health records.
Step 3 - Choose Data Sources
Quantitative sources:
- Structured questionnaires / validated scales
- Hospital records / electronic health records (EHR)
- Lab values, vital signs, imaging results
- Registries and administrative databases
Qualitative sources:
- Semi-structured interviews
- Focus group discussions
- Observation (field notes)
- Document analysis (clinical notes, discharge summaries)
Step 4 - Obtain Ethical Approval
- Submit to your Institutional Review Board (IRB) or Ethics Committee before any data collection.
- Prepare an informed consent form - written for patients, verbal for vulnerable populations.
- Address confidentiality, data storage, right to withdraw, and potential risks.
3. Designing the Questionnaire (Quantitative Component)
A. Types of Questions
| Type | Use Case | Example |
|---|---|---|
| Closed-ended (forced choice) | Measuring frequency, severity, demographics | "How many days did you experience pain this week?" |
| Likert scale | Attitudes, satisfaction, symptom burden | "Rate your pain: 1 (none) to 5 (severe)" |
| Visual Analogue Scale (VAS) | Continuous measurement (pain, mood) | 0-100 mm line |
| Dichotomous (Yes/No) | Screening, presence of a condition | "Do you have diabetes? Yes / No" |
| Multiple choice | Categorical variables | Comorbidities, medication types |
| Open-ended (free text) | Capturing nuance within a structured instrument | "Describe any other symptoms" |
B. Use Validated Instruments Where Possible
Avoid building from scratch if a validated tool exists. Examples:
- PHQ-9 - depression screening
- GAD-7 - anxiety
- SF-36 / SF-12 - general health-related quality of life
- AUDIT - alcohol use
- NRS (Numeric Rating Scale) - pain
- MRC Dyspnoea Scale - breathlessness
Validated tools have established reliability (Cronbach's alpha) and validity, saving time and allowing comparison with published data.
C. Questionnaire Structure
A well-structured clinical questionnaire typically follows this order:
- Cover page - study title, purpose, confidentiality statement, contact details
- Informed consent section (or attached separately)
- Socio-demographic data - age, sex, education, occupation, residence
- Clinical background - diagnosis, disease duration, comorbidities, medications
- Core outcome measures - validated scales or custom items matching your objectives
- Supplementary open-ended items - capturing anything not covered by closed questions
- Thank-you note and instructions for return
D. Writing Good Questions - Key Rules
- One idea per question - never double-barrelled ("Do you exercise and eat well?" - bad)
- Avoid leading questions - "Don't you agree that..." - bad
- Use plain language - aim for a Grade 6-8 reading level for patient-facing tools
- Avoid negatives - "I do not feel unwell" confuses respondents
- Pilot test - run the questionnaire with 5-10 people similar to your target population before full deployment; check for confusion, missing options, and time to complete
- Response options must be exhaustive and mutually exclusive
E. Questionnaire Modes of Administration
| Mode | Advantages | Disadvantages |
|---|---|---|
| Self-administered (paper) | Cheap, private | Literacy required, incomplete returns |
| Interviewer-administered | Clarifies confusion, higher completion | Interviewer bias, cost |
| Electronic (REDCap, SurveyMonkey, KoBoToolbox) | Automated, data export, skip logic | Requires device/internet access |
| Telephone | Reaches remote patients | No visual aids, response fatigue |
REDCap is the gold standard for clinical research - it supports skip logic, audit trails, e-consent, and secure HIPAA-compliant storage.
4. Designing the Qualitative Component
A. Semi-Structured Interview Guide
Unlike a questionnaire, an interview guide is a flexible framework, not a rigid script. It contains:
- Opening/rapport-building questions - "Can you tell me a little about your experience with [condition]?"
- Core topic questions - 5-8 open-ended questions aligned with your qualitative research questions
- Probes - follow-up prompts to deepen responses: "Can you tell me more about that?", "What did that feel like?", "Can you give me an example?"
- Closing question - "Is there anything else you would like to share that we haven't covered?"
B. Focus Groups vs. Individual Interviews
| Individual Interview | Focus Group | |
|---|---|---|
| Best for | Sensitive topics, detailed personal narratives | Shared experiences, generating diverse perspectives |
| Group size | 1:1 | 6-10 participants |
| Duration | 45-90 minutes | 60-120 minutes |
| Facilitator skill | Moderate | High (managing group dynamics) |
C. Qualitative Data Recording
- Always audio-record (with consent).
- Transcribe verbatim for thematic or content analysis.
- Keep field notes for non-verbal cues and context.
5. Integration - Mixing the Two Components
The key to mixed methods is how and when you integrate the data. Common strategies:
- Triangulation: Compare quantitative findings against qualitative themes - do they confirm, contradict, or complement each other?
- Explanation: Use interview findings to explain unexpected statistical results.
- Instrument development: Use qualitative themes to generate or refine questionnaire items for a second phase.
Document your integration decisions in your methodology section so reviewers can assess rigor.
6. Data Management and Quality Control
- Numbering and coding: Assign unique participant IDs; never use names in data files.
- Double data entry: For paper questionnaires, two people enter data independently and discrepancies are resolved.
- Missing data protocol: Define in advance how you will handle missing responses (exclusion threshold, imputation method).
- Version control: Label each version of your questionnaire (v1.0, v1.1) and never collect data with two different versions simultaneously without accounting for changes.
- Storage: Encrypted, password-protected files; physical paper stored in locked cabinets; retain for the period required by your institution (usually 5-10 years).
7. Summary Checklist Before Data Collection Begins
- Research objectives clearly defined
- Study design and timeline documented
- IRB/Ethics approval obtained
- Informed consent form prepared
- Sampling strategy and sample size calculated
- Questionnaire drafted and piloted
- Validated scales selected and permissions obtained (if required)
- Data collection mode confirmed (REDCap / paper / interview)
- Interview guide drafted and piloted
- Data management and storage plan in place
- Team trained on standardised data collection procedures
If you want me to go deeper on any specific section - such as building a sample questionnaire for a particular clinical condition, writing an interview guide, or setting up REDCap - just let me know.
This is a shared conversation. Sign in to Orris to start your own chat.