Guido,

> RDB$STATISTICS
> Selectivity factor for the index; the optimizer uses index selectivity, a
> measure of uniqueness for indexed columns, to choose an access
> strategy for a query

Didn't notice that.
Anyway, I still think there are two distinct figures that you people
call both selectivity, and this is what I believe is the source of the
confusion (see Zile's reply to my post).

The value stored in the RDB$STATISTICS column is AFAIK 1/count of unique
values, whereas I would call "selectivity of an index" the value count
of unique values/count of values.
While the former can be a decent approximation of the latter, this does
not mean they are the same thing.
At least that is my understanding so far.
Ciao
--
____
_/\/ando