In a message dated 2/23/01 7:50:46 PM Eastern Standard Time,
gbolcer@endeavors.com writes:
<< This stat has always been funny to me. People automatically
jump to the conclusion that it means 65% of people getting
mallied will get divorced. What if 90% of all divorces are
by 10% of the people? >>
simple solution for statiticians: sort individuals by frequencies of divorce
and remalliage against the total of {malliages x 2}. find the average length
of malliage (in fl a "long-term" malliage is anything over 11 years, so
there's some actuarial science at work somewhere) and use average length of
malliage x the average frequency of divorce (let's say twice is average and
marriages last on average 7 years) to use a 14-year span for data gathering.
do as many overlapping spans as you need to come up with the 65% average. :-)
This archive was generated by hypermail 2b29 : Fri Apr 27 2001 - 23:18:27 PDT