Ask Mallard
Core concept

When Mallard gives no sample size

4 min read

A missing headline sample size is a reason to inspect the sizing route, inputs and warnings before proceeding.

Fit

When it matters, and when it does not

It matters when

  • The plan gives no recruitment total or says a sizing contradiction blocks it.
  • The plan specifies simulation, a fixed design or a method outside the supported sizing catalog.

It matters less when

Before you start

What to check in your study

Decide these first
  1. The exact sizing route and its reasonSeparate missing inputs, a blocked headline, fixed-design power and a calculation that has not been run.
  2. The quantity and design to sizesummaryMeasuretargetclusteredframingCheck the summary measure, matching, clusters, prediction goal and any noninferiority margin.
It rests on these
  1. A proposed calculation is not a completed result.How to check: Targets and assumptions do not establish achieved power. A simulation result needs an actual run and its uncertainty.
  2. A substitute formula must size the same question.How to check: A paired continuous-means calculation does not size a matched case-control study.

Plain language

How it works

First inspect the reason. Missing required inputs stop a supported calculation. Generic simulation records a specification without executing it; none selects no calculator. Interrupted time series and stepped wedge report power for fixed designs, without a headline recruitment total.

How it works

Matched case-control and prediction-model development have no dedicated catalog form. Equivalence and unsupported noninferiority questions need a margin-based method or simulation. A narrow exception exists: two independent proportions with a risk-difference noninferiority margin. A superiority formula cannot replace it.

How it works

If a designated sizing contradiction blocks the headline, resolve the mismatch between the estimand and sizing method. Do not copy a number from another part of the plan.

The decision in view
Four reasons for no headline size: missing inputs, a blocked sizing contradiction, power for a fixed design, or a calculation not run. Each has a different next action.No headline: find the reason before choosing the next actionMissing inputsGather defensible inputsBlocked sizing contradictionResolve the mismatchFixed-design powerInterpret power for that designCalculation not runSpecify and run the right methodGeneric simulation and none supply no headline size. A fixed-design power result is a different answerfrom a recruitment total.
Conceptual diagram. It shows the structure of the idea, not data from any study.

Worked example

A worked example

Invented for illustration
Setting
A team plans a clustered study that needs bespoke simulation.
Question
Does writing a target risk reduction establish a sample size?
Approach
Assume a baseline risk of 10 in 100 and a target risk of 8 in 100.
Result
The assumed risks change from 10 to 8 in 100: a change of -2 percentage points.
What it means
That arithmetic defines the target contrast only. No simulation has been run, so it supplies neither achieved power nor a required sample size.

In practice

Recent studies PubMed search · not vetted

Searching PubMed…

Watch out

Common mistakes

The one to remember

No headline sample size does not mean that no planning is needed. Never replace an unsupported calculation with an unrelated formula just to obtain a number.

  • Reading a simulation specification as executed power.Instead: Run the specified design and analysis. Report successful fits, failures and Monte Carlo uncertainty.
  • Using an events-per-variable floor for prediction-model development.Instead: Use a method that addresses model performance, shrinkage and overall risk, such as Riley criteria.

Inside Mallard

How Mallard uses this

Generic simulation and none return no headline sample size. Mallard also withholds a headline for designated blocking sizing contradictions.

No Mallard code example points to this page.
Atlas node concept.no_sample_size, the id a plan cue records.

Related Learn pages

Related designs and methods

Evidence

Sources & further reading

  1. Minimum sample size for developing a multivariable prediction model: PART II - binary and time-to-event outcomes. (opens in a new tab) · Stat Med · doi 10.1002/sim.7992
  2. The size of a pilot study for a clinical trial should be calculated in relation to considerations of precision and efficiency. (opens in a new tab) · J Clin Epidemiol · doi 10.1016/j.jclinepi.2011.07.011
Prepared with AI assistance. Primary methodological and empirical source records were checked by PMID for title, authors, year and DOI on 2026-11-06. These sources support the general methods; tests pin the separate claims about Mallard. No independent professional approval is claimed.
Last source review: 2026-11-06
AI-generated educational content
Mallard uses reputable methodological and reporting sources, but this material may contain errors or omit important context. Verify important methodological decisions with the cited sources and, when appropriate, a qualified research methodologist or statistician.