I85PREP
⌘ K
ASEP Written Contest study materials
ASEP Written Contest study resources

ASEP Written Contest: Research Methods and Evidence-Based Practice

ASEP Written Contest · Work through your exam, one topic at a time.

All guide sections

Research Methods and Evidence-Based Practice

The Research Methods and Evidence-Based Practice domain tests whether you can appraise studies and apply evidence without overclaiming. Expect study-design questions, bias and validity concepts, diagnostic-test calculations, and interpretation of p-values and confidence intervals.

Core rules

  • Evidence hierarchy: systematic reviews and meta-analyses of randomized controlled trials rank above single RCTs, cohort studies, case-control studies, case reports, and expert opinion. The hierarchy is a starting point, not a substitute for critical appraisal.
  • Internal validity concerns whether the study’s design supports causal inference; external validity concerns generalizability.
  • Reliability is consistency; validity is accuracy. Random error reduces precision; systematic bias shifts results in one direction.
  • Confounding occurs when a third variable is associated with both exposure and outcome. Randomization, restriction, matching, and statistical adjustment address it.
  • p-values indicate the probability of data at least as extreme if the null hypothesis were true; they do not give the probability that the null is true. Confidence intervals show precision and effect size.
  • Diagnostic test metrics: sensitivity = TP/(TP+FN); specificity = TN/(TN+FP); PPV = TP/(TP+FP); NPV = TN/(TN+FN). PPV and NPV depend on prevalence.

Worked example

A screening test has 80% sensitivity and 90% specificity in a population with 10% prevalence. In 1,000 people: 100 have the condition; TP = 80, FN = 20. Of 900 without it, FP = 90, TN = 810. PPV = 80/(80+90) = 47.1%; NPV = 810/(810+20) = 97.6%. Even with good sensitivity and specificity, a positive result is a coin flip because prevalence is low. This is why screening tests need confirmation and why PPV cannot be transferred to a different prevalence setting.

Traps

  • Saying p = 0.05 means a 5% chance the null hypothesis is true.
  • Confusing sensitivity with PPV.
  • Assuming a test’s PPV is fixed across populations.
  • Treating statistical significance as clinical importance.
  • Confusing correlation with causation.
  • Ignoring dropout, blinding, and intention-to-treat analysis.
  • Overgeneralizing from a small or nonrepresentative sample.