Nursing Administration and Nursing Research
What this domain tests
Management theory and principles, nursing care delivery models, staffing and material management, quality assurance, and the research process from problem statement to statistics and ethics.
Core rules
- Classical contributors: Taylor (scientific management), Fayol (14 principles, unity of command), Weber (bureaucracy), Maslow and Herzberg (motivation and hygiene factors).
- Care delivery models: functional (task-based, efficient but fragmented), team nursing (team leader delegates), primary nursing (one nurse accountable for the whole plan), case method (one nurse, one patient).
- Delegation and supervision: delegate by competence and patient acuity, retain accountability, and never delegate assessment, care planning or evaluation to unlicensed staff.
- Research process: identify problem → literature review → hypothesis and objectives → design → sampling → data collection → analysis → interpretation → report.
- Designs: true experimental needs randomisation and a control group; quasi-experimental lacks randomisation; descriptive and correlational designs test no cause–effect claim.
- Errors: Type I error is rejecting a true null hypothesis (alpha level, usually 0.05); Type II error is accepting a false null hypothesis (beta). Power is 1 − beta.
- Statistics: chi-square for association between categorical variables; independent t-test for two group means; ANOVA for three or more group means; Pearson r for correlation.
Worked example
A study compares anxiety scores in three groups - yoga, music and routine care - measured on a standard scale. Reasoning: the dependent variable is interval and the independent variable has three independent groups, so one-way ANOVA with post-hoc comparison is chosen. A t-test would inflate Type I error across multiple pairwise comparisons.
Traps that cost marks
- Confusing the independent variable (the intervention) with the dependent variable (the outcome).
- Using repeated t-tests instead of ANOVA for three groups.
- Describing a study without randomisation as a true experiment.
- Swapping Type I and Type II errors, or treating p < 0.05 as proof of a large effect.