- Reproducible empirical experiments provide the strongest proof.
- Statistical evidence may be good, but tricky to handle. Even very bright and scientifically trained people may mishandle statistical data. A typical example is assuming that correlations mean cause and effect.
- First hand testimony may be good but people can be tricked either by the deceit of others or miscomprehension. If testimonial can be corroborated by others, then its value goes up, taking into account the risk of collusion and trickery.
- Allegory is a step away from direct experience and makes poorer evidence.
- Analogy is another move away from direct evidence and belongs in the realm of concept association and logic.
All of these types of evidence are present in politics. Unfortunately there may be little in the way of useful socialogical experiments and difficult to comprehend statistical evidence may be the best we can get. This points to the necessity for scientific governmental advisers in the Interactive Democracy system.