Comments on: D'Arrigo: Making Cherry Pie

By: Can’t see the Climate Forest For the Trees

Can’t see the Climate Forest For the Trees — Sat, 07 Jul 2012 15:09:49 +0000

[...] is frequently referred to as “cherry picking”. In an amazing comment about tree ring studies, Roseanne D’Arrigo said if you’re going to make cherry pie you have to pick cherries. The issue is also discussed in [...]

By: kim2ooo

kim2ooo — Thu, 17 May 2012 21:15:29 +0000

Reblogged this on Climate Ponderings.

By: Schmidt’s “Conspiracy Theory” « Climate Audit

Schmidt’s “Conspiracy Theory” « Climate Audit — Thu, 17 May 2012 21:03:23 +0000

[...] an astonished panel that “you had to pick cherries if you want to make cherry pie” (see here). Although this evidence was highly relevant to the subject of the NAS inquiry and occasioned a [...]

By: Paleoclimate – Rotten to the core « the Air Vent

Paleoclimate – Rotten to the core « the Air Vent — Sat, 26 Nov 2011 13:40:02 +0000

[...] need to pick cherries to make cherry pie”, D’Arrigo NAS panel [...]

By: ATI’s Chesser in American Spectator: Michael Mann Acts

ATI’s Chesser in American Spectator: Michael Mann Acts — Wed, 14 Sep 2011 12:59:02 +0000

[...] emails not relevant to valid scientific concerns.” Plucky, D’Arrigo is famous for explaining to the National Academy of Sciencesthat “cherry picking” is necessary if you want to make cherry pie. And as for what she [...]

By: The American Spectator : AmSpecBlog : Mann Acts

The American Spectator : AmSpecBlog : Mann Acts — Tue, 13 Sep 2011 21:45:39 +0000

[...] are personal emails not relevant to valid scientific concerns." Plucky, D'Arrigo is famous for explaining to the National Academy of Sciences that "cherry picking" is necessary if you want to make cherry pie. And as for what she thought of [...]

By: John A

John A — Sun, 12 Mar 2006 23:02:37 +0000

Re #45

Wouldn’t it be fascinating to actually give a large set of observations for a particular but unknown locale and ask these people to reproduce a temperature record?

They can pass a double blinded test, then I’ll be impressed.

By: Lubo Motl

Lubo Motl — Sun, 12 Mar 2006 22:18:13 +0000

Dear Steve,

“cargo cult sciences” is a term close to my heart. Yes, when we make the assumption that *every* new failure, discrepancy, or divergence is a reflection of a new kind mechanism, then we are surely doing cargo cult science.

All the best
Lubos

By: jae

jae — Fri, 10 Mar 2006 02:33:04 +0000

If you are making cherry pie, none of this matters. It is simply terrible science. You can’t hand-select data that matches your theory. PERIOD.

By: Paul Penrose

Paul Penrose — Fri, 10 Mar 2006 01:23:31 +0000

I have been thinking of the r^2 versus RE issue and I have come to this conclusion: It’s possible that under these conditions the RE statistic is somewhat more indicative of skill than the r^2. In this case if you had a weak r^2, a strong RE could still compensate for it. However, I don’t see how anybody could support the idea that the r^2 statistic is completely useless to the point of not even computing/reporting it. I just can’t think of any legitmate situation where you could justify totally discounting the r^2 statistic.