[Same data - different conclusions: Analytical flexibility in medical research can be considerable]. / Olika slutsatser från samma data.
Lakartidningen
; 1202023 05 15.
Article
in Sv
| MEDLINE
| ID: mdl-37191395
ABSTRACT
Analysis of research data entails many choices. As a result, a space of different analytical strategies is open to researchers. Different justifiable analyses may not give similar results. The method of multiple analysts is a way to study the analytical flexibility and behaviour of researchers under naturalistic conditions, as part of the field known as metascience. Analytical flexibility and risks of bias can be counteracted by open data sharing, pre-registration of analysis plans, and registration of clinical trials in trial registers. These measures are particularly important for retrospective studies where analytical flexibility can be greatest, although pre-registration is less useful in this context. Synthetic datasets can be an alternative to pre-registration when used to decide what analyses should be conducted on real datasets by independent parties. All these strategies help build trustworthiness in scientific reports, and improve the reliability of research findings.
Search on Google
Collection:
01-internacional
Database:
MEDLINE
Main subject:
Biomedical Research
Type of study:
Observational_studies
/
Risk_factors_studies
Limits:
Humans
Language:
Sv
Journal:
Lakartidningen
Year:
2023
Document type:
Article