Re: observing intercourse

["John M Price, PhD" <jmprice@xxxxxxxxxx>]

On Wed, 24 Nov 1999, Brian Morgan (02920 222656) wrote:

> Can you explain why harm may be present in clinical studies but not when more
> general survey work is done?

The short answer is sample bias.  It is similar to commenting on the
health of a community by looking only at those folk in a hospital.  Not a
good sample.

> Is harm not really there are at all, and what is wrong with the clinical study
> if it shows there is harm?

My personal guess, based on a global view, is that there is no real harm
involved.  Rather, the harm is after the fact - things being blown out of
proportion. 

The other problem is confirmatory bias.  Here clinicians believe that this
would cause harm, find a kid with a problem and learn the kid saw the
'primal scene' then attribute the problem to that event using the logical
error of /post hoc ergo propter hoc/ ({problem} was after the fact,
therefore because of the fact).

Finally, as far as I know, only psychodynamic folk think this is a problem
anyway.

> If there is a discrepancy between clinical measures of harm and more general
> surveys in this topic then what about other topics?

See not only the paper itself, but also the huge controversy surrounding
the Rind et al paper on college students who experienced sexual contact
(their preferred term for some of those events) as kids.  I have a copy of
the presentation version (not the full paper) on my FTP site at:

ftp://ftp.calweb.com/users/j/jmprice/w7.pdf

The main point that seems lost on lots of folk is that to determine what
child abuse of any form causes, you have to study children who were
abused.  You can't simply select those who have demonstrated problems.
Finkelhor (and Boney-McCoy?) showed that one of the best predictors of
psychopatholgy following an abusive event is previous psychopathology.  So
again, you have to be sure to ascertain, to the extent possible, that the
abused and nonabused groups are equivalent prior to the abusive event, as
intrasubject variables might be confounding the uncontrolled view.

Hope that helps.

Enjoy the holiday.

[snip old mail]
-- 
John M. Price, PhD                                     jmprice@calweb.com
Life: Chemistry, but with feeling!      |      PGP Key on request or FTP!
Comoderator: sci.psychology.psychotherapy.moderated          Atheist# 683