A measure of the evidence provided by the data, D, in favour of model M1 as opposed to model M2. The Bayes factor is B, given by
The value of B may be assessed with the following table, derived from that suggested by Jeffreys:
2ln(B)  | evidence in favour of M1  | 
|---|
<0  | Negative  | 
0–2.2  | Not worth more than a bare mention  | 
2.2–6  | Positive  | 
6–10  | Strong  | 
>10  | Very strong  |