Quantitative vs. Qualitative
Quantitative: numerical data, objective measurement, statistical analysis, generalizability. Answers "what" and "how much" questions. Uses structured tools, large samples, controlled settings. Qualitative: narrative data, subjective experience, thematic analysis, depth over breadth. Answers "why" and "how" questions. Uses interviews, observations, focus groups, small samples. Mixed methods combine both for comprehensive understanding. Each approach has strengths — choose based on the research question.
Experimental Designs
RCT (Randomized Controlled Trial): gold standard for treatment efficacy. Random assignment to intervention vs. control. Eliminates selection bias, controls confounders. Blinding (single, double, triple) reduces bias. Limitations: ethical constraints, cost, external validity. Quasi-experimental: no randomization (e.g., before-after study on one unit). Weaker internal validity but more feasible in real-world settings. Pre-experimental: one-group, post-test only — weakest design.
Observational Designs
Cohort: follows a group over time (prospective) or uses records (retrospective). Measures incidence and risk. Case-control: compares people with condition (cases) to those without (controls). Good for rare diseases. Cross-sectional: data at one point in time. Measures prevalence. Quick and inexpensive but cannot establish causation. Case series/case reports: clinical description of interesting cases. Generates hypotheses but no statistical conclusion.
Qualitative Approaches
Phenomenology: explores lived experience of a phenomenon (e.g., "What is the experience of being a family member during end-of-life care?"). Grounded theory: develops a theory from data (e.g., "How do nurses make decisions during a code?"). Ethnography: studies cultural patterns (e.g., nursing culture on a unit). Narrative: individual stories. Qualitative data collection: semi-structured interviews (most common), focus groups, observation, document analysis. Analysis: coding, themes, constant comparison. Trustworthiness: credibility, transferability, dependability, confirmability.
Sampling
Quantitative: probability sampling (random, stratified, cluster, systematic) — every member has known probability of selection. Enables generalizability. Non-probability (convenience, purposive) — faster but limited generalizability. Power analysis determines sample size. Qualitative: purposive sampling (select participants with rich experience), theoretical sampling (sample based on emerging theory). Sample size determined by data saturation — when no new themes emerge. Typically 10-30 participants for interviews.
Important Notes
- The research question determines the design — you cannot use an RCT for a question about lived experience.
- Convenience sampling (the most common in nursing research) limits generalizability — acknowledge this limitation.
Clinical Tips
- Hierarchy of evidence: meta-analysis > RCT > cohort > case-control > cross-sectional > case report > expert opinion.
- Qualitative research is NOT "lesser" than quantitative — it answers different questions. Both are needed for evidence-based practice.
Summary
Understanding research methods enables nurses to critically appraise studies and apply findings to practice. Quantitative designs answer questions about effectiveness; qualitative designs answer questions about experience. Choose the design that matches the question.
