Skip to content
ASCIENDE

The Asciende advantage

Free Mexico shipping from $2,500 MXN

Shipping details

Research concepts

What a Well-Designed Negative Result Shows

Negative results: Which effect could this experiment have detected?

Source editorial review:

Before interpreting the result

Which effect could this experiment have detected?

  • Examine method sensitivity and control behavior.
  • Record the estimate, interval and tested conditions, not only the p-value.
  • Separate absence of evidence, evidence of equivalence and assay failure.

Negative results

A negative result is interpretable when the assay could detect the relevant phenomenon and the design bounds the conclusion. Absent signal can reflect absence under those conditions, inadequate sensitivity or system problems. Controls, uncertainty and measurement scope help distinguish these possibilities.

A reading case: what to check

When summarizing a study with no detected difference, retain the system, variable and range it could assess. Avoid writing 'has no effect' without those bounds. With functioning controls and an adequate design, the result may usefully constrain a hypothesis. If controls are missing, that uncertainty also belongs in the conclusion.

What to preserve in the record

A negative result delimits a claim when design and precision are adequate. It does not establish that the phenomenon is impossible in every model or condition.

Questions and answers

Which effect could this experiment have detected?

A negative result is interpretable when the assay could detect the relevant phenomenon and the design bounds the conclusion. Absent signal can reflect absence under those conditions, inadequate sensitivity or system problems. Controls, uncertainty and measurement scope help distinguish these possibilities.

What should the review record preserve?

A negative result delimits a claim when design and precision are adequate. It does not establish that the phenomenon is impossible in every model or condition.

Sources

  1. Using Bayes to get the most out of non-significant results.
  2. Equivalence Tests: A Practical Primer for t Tests, Correlations, and Meta-Analyses.
  3. ARRIVE 2.0: explicación de la justificación del tamaño de muestra
Ayuda