Ask Mallard

Learn

Press / to jump to the search box.

Learning paths

A few pages read in order, for when you would rather be walked through a topic than search it.
  • Plan an observational study from existing records
    For a chart review or registry study, from the question to the time-to-event model.
    7 pages · about 26 min
  • Design a randomized trial
    For a first trial protocol, from what is being estimated to the prespecified analysis.
    7 pages · about 26 min
  • Make a causal claim carefully
    For anyone asking whether an exposure causes an outcome without randomizing it.
    7 pages · about 24 min
  • Build and check a prediction model
    For a risk score or prediction tool, from development to validation.
    5 pages · about 18 min
  • Evaluate a change on your own ward
    For a quality improvement project, and for telling a real change from noise.
    6 pages · about 22 min

Core research concepts

The ideas that decide whether a result can be believed.
Complete cases or multiple imputation?
Choose a missing-data strategy from the mechanism and estimand, not a percentage rule.
4 min read
Prediction leakage, overfitting and optimism
Keep information boundaries intact when developing and evaluating a prediction model.
4 min read
From research question to estimand
Turn a broad clinical idea into a defined target quantity before selecting the study design and analysis.
4 min read
Time zero, index date and follow-up
Align eligibility, exposure assignment and the start of outcome follow-up so the comparison answers the intended question.
4 min read
Confounding
When groups differ in causes of the outcome, the comparison can mislead.
3 min read
Effect modification and interaction
When the size of an effect differs across levels of another variable.
3 min read

Study designs

How a study is built so its answer means something.
Cohort or case-control?
Choose how to sample people and observations for an etiologic question.
4 min read
Developing and validating a measurement instrument
Establish what a score measures, for whom, and how consistently it does so.
4 min read
Scoping reviews and evidence maps
Map what evidence exists and how it is organized before making claims about effects.
4 min read
Quality improvement studies
Study a change in care delivery using explicit measures, context and a comparison suited to the claim.
4 min read
Survey and questionnaire studies
Plan who can respond, what the questions measure and how responses support the intended inference.
4 min read
Randomized controlled trial
Assigns participants to groups by chance, so baseline differences are balanced on average.
4 min read

Statistical analyses

What each model estimates, and what it assumes to do it.
Logistic regression or Cox regression?
Distinguish an outcome at a fixed horizon from the timing of that outcome.
4 min read
Mixed effects or GEE?
Match the interpretation of a correlated-outcome model to the question.
4 min read
Propensity matching or weighting?
Choose the target population before choosing how to balance observed covariates.
4 min read
Interrupted time series and segmented regression
Estimate a prespecified change in an outcome trajectory after an intervention.
4 min read
Complex survey analysis
Estimate population quantities while respecting sampling weights, strata, clusters and subgroup membership.
4 min read
Prediction model performance and validation
Evaluate calibration, discrimination and usefulness with a validation strategy that protects against leakage and optimism.
4 min read

Glossary

79 terms in plain language, for the jargon a plan and the literature use without stopping to explain.
Open the glossary →