Re: speed of light
Bill Hamilton (hamilton@predator.cs.gmr.com)
Fri, 12 Nov 1999 13:05:50 -0500
At 02:20 PM 11/12/99 +0000, Graham Richard Pointer wrote:
>
>There is a response to Setterfield's arguments (and indeed those of many
>YEC's) in Alan Hayward's book "Creation and Evolution".
>
>The main argument is basically that Setterfield invented data to fit his
>theory e.g Ole Romer's calculation of the speed of light from eclipses of
>Jupiter's moons gave a figure less than the currently accepted version, but
>Setterfield gives a value far higher than the current one- which
>conveniently fits his graph.
>
Thanks for posting this, Graham. I wasn't aware that Setterfield had
cooked the data. As I have pointed out before, Setterfield's paper is one
of the most abysmal examples I have ever seen of bad statistical inference.
He took the data (which he cooked in the first place) and then looked
through tables of mathematical functions until he found one that fit the
data exactly. (Actually, come to think of it, he probably "adjusted"
Roemer's data until it fit the curve he wanted to use). The insistence
that the data ought to fit the curve exactly in itself ought to be a huge
red flag to any reader who knows the first thing about statistical inference.
I have heard that an employee of SRI somehow managed to get one or both of
Setterfield's papers released as SRI reports, and that the employee was
dismissed. Does anyone know whether this is a true story?
Bill Hamilton
----------------------------------------------------------------------------
William E. Hamilton, Jr., Ph.D.
Staff Research Engineer
Electrical and Controls Integration MC 480-106-390
GM R&D Center
30500 Mound Road
Warren, MI
hamilton@predator.cs.gmr.com / whamilto@mich.com (home)