After some prompting from a recent summary post from Glenn Reynolds,
John Lott has posted some clarifications. John admitted that the coding
errors that Ayres and Donohue pointed out do exist.

Before posting to this list I checked Tim Lambert's weblog to see what
his take on this was. Though accepting the fact that Lott has admitted
the errors exists he states that Lott doesn't address what the
implications for his results are. I have a hunch that that was posted by
Tim before Dr. Reynolds posted an update to his most recent Lott Post.
In that update Dr. Reynolds links to this:

http://www.johnlott.org/files/statement.html

The second paragraph seems to be Dr. Lott's statement on the impact
those coding errors have on the results. Dr. Lott states that the errors
have been corrected in the downloadable data sets on his website for
those inclined and adept at confirming his statements.

For those of you interested, the two posts from Dr Reynolds in
chronological order are:
http://www.instapundit.com/archives/011099.php
http://www.instapundit.com/archives/011118.php

Tim's post referred to above:
http://cgi.cse.unsw.edu.au/~lambert/cgi-bin/blog/2003/08#0821

-Gene Hoffman

Reply via email to