Fabrication and Falsification

 

Fabrication involves inventing data, observations, participants, experiments, results, or other research information that did not exist. Falsification involves manipulating research materials, procedures, data, images, or results so that the research record no longer accurately represents what was conducted or observed.

Examples of Fabrication
  • Inventing experimental observations or survey responses.
  • Creating datasets that were not actually collected.
  • Reporting experiments that were never conducted.
  • Inventing participants or research subjects.
  • Creating unsupported numerical results.
Examples of Falsification
  • Changing data to produce a desired outcome.
  • Removing observations without a valid methodological reason.
  • Manipulating figures or images in a misleading manner.
  • Selectively reporting results while concealing contradictory findings.
  • Misrepresenting statistical or computational analysis.
Research Integrity

Authors should retain appropriate research records and report analytical decisions transparently. Legitimate data cleaning, preprocessing, exclusion criteria, transformation, or statistical procedures should be documented sufficiently to allow readers to understand the research process.