Need help!
Message posted by Don on November 5, 1999 at 12:00 AM (ET)

I have a collection of data points
and "factoids" that I do not know
how to summarize. This collection is
composed of little pieces of information on
membership in an organization. The bits
of data are in the form of "Of the people in
a single given meeting of the organization, one third
have belonged to the organization less than one year, one third have
belonged to the organization from two to five years, and
and one-third have belonged to the organization more than five
years.

If a person drops out and returns to the organization and to the meetings, they are counted
as a fresh new member as if they had never
been a member before.

There are several other little bits I would
be happy to provide if anyone knows how
to set the problem up and solve it.

I am interested in arriving at a curve or formula that
will tell me what percentage of new members last any given
amount of time in the organization.

I also suspect, but don't know, that the drop out rate
may not be uniform over time, and may have
ups and downs at particular points.

I would also like to know if it is
basically true that given that a person
has been a member for x days, it follows
that the odds of his remaining a member for
x + 1 days, x + 2 days, and so on steadily
increase with the passage of time.

Is this an appropriate problem to post?

Thank you.

READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: Need help!
Message posted by jg on November 5, 1999 at 12:00 AM (ET)