I’m no journalist, so I don’t have a firm grip on the evidentiary standards of the profession. Political scientists would tend to look on defector-provided information as inherently suspect; defectors are not randomly selected individuals, and they likely harbor particular hostility toward their home regime. Moreover, exiles and defectors at best have only a partial understanding of the situation in their home countries. The longer that exiles have been away, the less useful information they can give.
The CIA knows all this, which is why it tends to discount defector testimony unless that testimony is backed up through other sources. Why the journalistic community couldn’t wrap its mind around this in 2002 is beyond me.