STRATA
Skip to study
XLVII

I.05 / GNOSIS

Meta-science

Which findings survive replication, and which features of the original study account for the difference?

A source-led introduction1 source references0 related experiments

Start with the question.

Meta-science studies the machinery by which findings are produced, checked, circulated, and remembered. The Open Science Collaboration's coordinated psychology project offers a documented example of examining replication across a selected group of studies. Its value here is methodological: the unit of analysis becomes a research process, including sampling, analysis choices, access to materials, and the criteria used to call two results consistent.

Inspect a finding's route into a summary.

Create a table for five papers on one question. Record the planned outcome, reported outcome, sample size, access to materials, and whether an independent follow-up exists. Keep an absent record as missing information rather than silently treating it as a failed experiment.

What would you inspect?

  • The denominator used for any reported success or replication rate.
  • Differences between an effect estimate, its uncertainty, and a significance threshold.

Take the next step.

Select one field-specific claim about reliability and reconstruct the studies included and excluded before comparing its reported rate with another survey.

Read the supporting material.

  1. Estimating the reproducibility of psychological science

    Open Science Collaboration / Science · A coordinated replication project and its stated design; supports study-level reproducibility questions.

Continue through the versioned study records.