Skip to main content

Scientific Standards and Quality Assurance

Science has developed various mechanisms to ensure the quality and reliability of its results.

The Peer Review Process​

The peer review process (assessment by fellow experts) is a central quality assurance mechanism in science.

The peer review procedure:

  1. Submission: Researchers submit their work to a scientific journal.
  2. Editorial pre-screening: An editor checks whether the work meets basic criteria.
  3. Review: The work is sent to several independent experts (peers).
  4. Evaluation: The reviewers assess the work with regard to methodology, results, interpretation and significance.
  5. Decision: Based on the reviews, the editor decides on acceptance, revision or rejection.
  6. Revision: If necessary, the authors revise their work in accordance with the reviewers' comments.
  7. Publication: After acceptance, the work is published.

Strengths of peer review:

  • Quality control by experts
  • Identification of errors or weaknesses
  • Improvement of clarity and precision
  • A filter against pseudoscientific or methodologically weak work

Weaknesses of peer review:

  • Possible bias on the part of the reviewers
  • The time required and resulting delays
  • A conservative tendency (favouring established ideas)
  • A limited ability to detect fraud

Replication and Reproducibility​

Replication (repeating a study in order to confirm its results) and reproducibility (the ability to obtain the same results using the same methods) are fundamental principles of science.

The importance of replication and reproducibility:

  • Confirmation of the reliability of results
  • Identification of chance findings or errors
  • Testing of generalisability under various conditions
  • Detection of fraud or questionable research practices

The replication crisis: In recent years, various disciplines, particularly psychology and biomedicine, have experienced a "replication crisis", in which many published results could not be replicated.

Causes of the replication crisis:

  • Publication bias (the tendency to publish only positive results)
  • "p-hacking" (selective analysis in order to obtain statistically significant results)
  • Insufficient statistical power
  • A lack of transparency regarding methods and data
  • Career incentives that place quantity above quality

Approaches to a solution:

  • Preregistration of studies
  • Open science practices (open data, open methods)
  • Improved statistical methods
  • Appreciation of replication studies
  • Changing the incentive structures in science

Scientific Integrity​

Scientific integrity refers to adherence to ethical principles and professional standards in research.

Basic principles of scientific integrity:

  • Honesty: Truthful presentation of methods, data and results
  • Objectivity: Minimisation of distortions and bias
  • Care: Accurate and conscientious conduct of research
  • Openness: Willingness to share data, methods and results
  • Respect: Ethical treatment of research participants, colleagues and the environment
  • Accountability: Accountability to the scientific community and to society

Scientific misconduct:

  • Fabrication: Inventing data or results
  • Falsification: Manipulating research materials, equipment or processes
  • Plagiarism: Using the ideas or words of others without appropriate acknowledgement
  • Self-plagiarism: Reusing one's own work without appropriate citation
  • Questionable research practices: Practices that, while not regarded as misconduct, undermine the integrity of research

Example: The case of the South Korean researcher Hwang Woo-suk, who in 2004-2005 published groundbreaking results in the field of cloning human embryos that later turned out to be fabricated.