Evidence Grading¶

Information on racing pigeon genetics and health comes from a wide range of sources, and its evidential strength is uneven. From rigorous peer-reviewed studies to scattered fancier experience, credibility differs enormously. Understanding an evidence grading system is the basis for reading information scientifically and deciding rationally.

Plain-language explanation (an analogy)¶

Evidence grading is like giving a piece of information a credit rating - not every claim is equally trustworthy. A rigorous peer-reviewed study is like a "bank certificate of deposit"; experience passed on by word of mouth among fanciers is like "hearsay". Learning to read the "rating" is what lets you pick out the genuinely sound conclusions from a mass of information.

Why evidence grading is needed¶

The racing pigeon field contains a great deal of "experiential" and "legendary" information, among which:

some information has no scientific basis;
some information is individual experience that has never been systematically verified;
some information is exaggerated or misleading;
conclusions that have actually been verified by rigorous research are relatively limited.

Evidence grading helps us tell credible information apart within a mass of claims and reach more rational judgements.

The evidence level system¶

A simplified evidence grading framework (from high to low):

Systematic reviews and meta-analyses: combine several high-quality studies; the strongest evidence;
Randomised controlled trials: control variables and set up controls; stronger causal inference;
Cohort and case-control studies: observational studies that can suggest associations;
Cross-sectional studies and case reports: provide clues, with limited evidential strength;
Expert opinion and experience: has reference value but lacks systematic verification.

Association versus causation¶

A key concept when reading genetic association studies is that association is not causation:

Association: a genotype and a trait occur together statistically;
Causation: the genotype directly causes the change in the trait;
many gene association studies can only suggest a "possible association" and cannot establish causation;
confounding factors, multiple testing and insufficient sample size can all produce spurious associations.

Principles for reading research conclusions¶

When faced with research conclusions about racing pigeons, follow these principles:

Check the source: does it come from a peer-reviewed academic journal;
Check the method: is the study design rigorous and the sample size adequate;
Check the wording: is the conclusion cautious, does it acknowledge its limitations;
Check reproducibility: has the conclusion been verified repeatedly by more than one study;
Stay cautious: be wary of wording such as "absolute", "guaranteed" or "certain".

This site's position on evidence¶

This research center holds to the following positions:

cite only real, traceable literature;
use cautious wording for research conclusions (such as "research suggests", "may be associated");
keep "scientific evidence" and "experiential reference" clearly apart;
do not exaggerate, do not fabricate, do not mislead.

Main site testing services¶

To learn more about racing pigeon genetic testing services, visit the Seno Genetic Testing Center main site -> Knowledge Center.