What exactly is ‘N’ in cell culture and animal experiments?
Key Points
- To evaluate how sample size (N) and independent replication are defined in animal experiments and to quantify the prevalence of pseudoreplication in published literature.
- Surveyed a random sample of published animal experiments from 2011 to 2016.
- Analyzed study designs in which interventions were administered to parents and outcomes were evaluated in their offspring.
- Only 22% of surveyed studies (95% CI = 17%–29%) correctly replicated the appropriate entity-intervention pair to allow valid statistical inference.
- Pseudoreplication occurred in 46% of studies (95% CI = 38%–53%), artificially inflating sample size and false-positive rates.
- A total of 32% of studies (95% CI = 26%–39%) provided insufficient experimental information to evaluate replication validity.
Authors
Discussion
No takes yet. Share an insight, caveat, or question.
Implication
Systematic survey finds widespread pseudoreplication in published animal research, highlighting severe inflation of statistical significance and false positives across preclinical literature.
Cite This Study
Lazic et al. (2018) studied this question.