Archived material Access restricted Article
Rantburg

Today's Front Page   View All of Thu 06/25/2009 View Wed 06/24/2009 View Tue 06/23/2009 View Mon 06/22/2009 View Sun 06/21/2009 View Sat 06/20/2009 View Fri 06/19/2009
1
2009-06-25 Syria-Lebanon-Iran
Statistics hint at fraud in Iranian election
Archived material is restricted to Rantburg regulars and members. If you need access email fred.pruitt=at=gmail.com with your nick to be added to the members list. There is no charge to join Rantburg as a member.
Posted by Lumpy Angaith3743 2009-06-25 01:55|| || Front Page|| [1 views ]  Top

#1 I guess statisticians have to make a living too.
Posted by JohnQC 2009-06-25 09:27||   2009-06-25 09:27|| Front Page Top

#2 The rest of us, particularly at Rantburg and the people in Iran already knew.
Posted by JohnQC 2009-06-25 09:28||   2009-06-25 09:28|| Front Page Top

#3 it is good to have professional statisticians weigh in here

back when I was in my 20s and 30s, people thought statisticians would be prominent players in policy analysis because it was important to have accuracy and precision before committing to policy

however, it didn't work out that way; it turned out politicians (both Rs and Ds) and the suck up types that depend on the favor of the politicians resent actual scientific analysis. (I got in trouble a few times for showing policy makers flaws in the p.c. analysis of the times).

I particularly like Silver's analysis because a lot of bloggers (mostly followers of Andrew Sullivan) were touting the constant proportion effect as mathematical certainty of fraud.
Posted by Lord garth 2009-06-25 10:13||   2009-06-25 10:13|| Front Page Top

#4 maths alone can't prove fraud.

But statistics are the superior technique for taking a census to determine representation, right?
Posted by Glenmore 2009-06-25 10:23||   2009-06-25 10:23|| Front Page Top

#5 maths alone can't prove fraud.

Prove, no. But very strongly indicate, absolutely. Just like Einstein's theories were only hypotheses until the astronomers saw starlight curved around the solar eclipse. Thank you for the perspective, Lord garth.
Posted by trailing wife ">trailing wife  2009-06-25 13:41||   2009-06-25 13:41|| Front Page Top

#6 the argument against using statistics to determine a census is constitutional, not scientific

the Constitution (Art 1. Sec 2) says 'enumeration' which in that century presumably meant 'a counting'
Posted by Lord garth 2009-06-25 13:44||   2009-06-25 13:44|| Front Page Top

23:59 newc
23:59 Atomic Conspiracy
23:44 Popnsyoa
23:31 OldSpook
23:29 OldSpook
23:20 OldSpook
23:14 Sockpuppet of Doom
22:41 Pappy
22:40 Frank G
22:35 Pappy
22:24 tu3031
22:09 Richard of Oregon
22:07 3dc
21:55 Muggsy Glink
21:48 Herman Thatch1710
21:42 Barbara Skolaut
21:40 Barbara Skolaut
21:38 Barbara Skolaut
21:32 Barbara Skolaut
21:01 Frank G
20:58 3dc
20:57 Butthead
20:52 Frank G
20:41 Frank G









Paypal:
Google
Search WWW Search rantburg.com