Date: Fri, 2 Apr 2004 00:30:30 -0500 (EST)
From: "Keith F. Lynch" <kfl at KeithLynch.net>
To: WSFA members <WSFAlist at KeithLynch.net>
Subject: [WSFA] Re: Washington Slush Fund Association (WSFA)?
Reply-To: WSFA members <WSFAlist at keithlynch.net>
Sam Lubell wrote:
> Keith, you just did this analysis, what's the typical non-overlap
> between Virginia and Maryland meetings?
In the report I emailed you, I said:
Over the past year, the two meetings each month averaged 15 members
in common (15 people who attended one meeting attended the next),
and averaged 41 different people total (41 distinct people attended
one or both). The attendance overlap between consecutive meetings
in the same location was 17, only slightly higher than that between
consecutive meetings in different locations.
But it occurs to me that that answers the wrong question. We should
really only be looking at WSFA members. Some of the non-overlap is
due to people who only attend one meeting. On average, each meeting
gets 1.2 people who only attend one meeting, and there's obviously no
overlap between their attending Virginia and Maryland meetings. And
they can't vote in our elections anyway. Nor can the 0.4 people at each
meeting who show up for exactly two meeting and are never seen again.
I don't know who is a paid up WSFA member, but I do know who has
attended three or more meetings. Counting just those people, in the
past two and a quarter years, ignoring the two months in which there
was only one meeting in which attendance was taken, the attendance
has been:
Average Median Lowest Highest
All meetings: 29 29 13 42
First Friday: 29 31 15 42
Third Friday: 28 28 13 39
Both meetings in a month:
intersection: 15 16 7 24
union: 41 42 33 51
att. first, but not third: 14 15 2 33
att. third, but not first: 13 13 4 30
Consecutive First Fridays:
intersection: 18 19 9 29
union: 40 40 30 49
Consecutive Third Fridays:
intersection: 16 16 9 24
union: 41 40 31 50
("Intersection" means the number of people each of whom were at both
meetings. "Union" means the number of people each of whom were at
either or both meetings.)
Why two and a quarter years? Because that's a convenient width on my
80 column screen for charts with one column per meeting and room left
over for the person's name. And it begins at the start of a year.
So I had already done some calculations based on that span of time.
It also seems to be a good compromise between so little time that the
data is swamped by brief fluctuations, and so long a time that WSFA
has changed significantly since it began.
--
Keith F. Lynch - http://keithlynch.net/
Please see http://keithlynch.net/email.html before emailing me.